Company Transport and Logistics Center 3 V

COMMUNICATIONS

OPERATION AND ECONOMICS INCREASING THE FUNCTIONING EFFICIENCY OF THE WORKING WAREHOUSE OF THE “UVK UKRAINE” COMPANY TRANSPORT AND LOGISTICS CENTER 3 V. Aulin, A. Hrynkiv, O. Lyashuk, Y. Vovk, S. Lysenko, D. Holub, T. Zamota, A. Pankov, M. Sokol, V. Ratynskyi, O. Lavrentieva SIMULATION OF A QUEUING SYSTEM OF A POST OFFICE IN ANYLOGIC SOFTWARE 15 S. Dutkova, K. Achimsky, P. Drozdziel ACCURACY ASSESSMENT OF AIRCRAFT POSITIONING USING THE DUAL-FREQUENCY GPS CODE OBSERVATIONS IN AVIATION 23 K. Krasuski, S. Savchuk DESIGNING OF PARKING SPACES TAKING INTO ACCOUNT THE PARAMETERS OF DESIGN VEHICLES IN RUSSIA 31 M. Mikusova, J. Abdunazarov, J. Zukowska, A. Usmankulov DETERMINANTS OF LABOR MARKET IN JAKARTA METROPOLITAN AREA: A SURVIVAL ANALYSIS OF COMMUTERS 42 I. W. Wardhana, D. Setyawan, Khairunnisah METHODS OF PREDICTING THE HEADING, PITCH AND ROLL ANGLES FOR AN UNMANNED AERIAL VEHICLE 52 D. Wierzbicki, K. Krasuski ANALYSIS OF ROAD SAFETY IN POLAND AFTER ACCESSION TO THE EUROPEAN UNION 60 R. S. Jurecki

MECHANICAL ENGINEERING OPTIMIZATION PROCEDURE OF ROLLER ELEMENTS GEOMETRY WITH REGARD TO DURABILITY OF SPHERICAL ROLLER BEARINGS 68 J. Steininger, S. Medvecky, R. Kohar, T. Capak

ELECTRICAL ENGINEERING BASIC COMPARISON AND EVALUATION OF FUNCTIONALITY AC-AC MATRIX CONVERTER CONCEPTS FOR HEV VEHICLE - PART II 73 B. Dobrucky, S. Kascak, M. Prazenica, R. Konarik ASSESSMENT OF SECURITY RISKS IN RAILWAY TRANSPORT USING THE FUZZY LOGICAL DEDUCTION METHOD 79 L. Hadacek, L. Sivakova, R. Sousek, M. Zeegers CIVIL ENGINEERING THE CONTINUOUS MONITORING OF SELECTED RAILWAY STRUCTURES USING THE AUTONOMOUS DATA LOGGER 88 D. Janostik, V. Nohal, H. Seelmann, J. Smutny

MANAGEMENT SCIENCE AND INFORMATICS ANALYSIS OF THE ACCEPTANCE FACTOR OF ANDROID-BASED PARKING INFORMATION SYSTEMS IN INDONESIA 97 R. Ilham, N. Shonhadji, H. Yutanto, D. Ekaningtyas APPLICATION OF SIMULATION METHODS FOR STUDY ON AVAILABILITY OF ONE-AISLE MACHINE ORDER PICKING PROCESS 107 M. Kostrzewski, J. Gnap, P. Varjan, M. Likos FORMATION AND OPTIMIZATION OF TRANSPORT NETWORKS IN THE CONDITIONS OF THE DIGITAL INTEGRAL ENVIRONMENT FOR TRANSPORTATION 115 B. Zhelenkov, I. Safonova, Y. Goldovsky, E. I. Dmitrieva

SAFETY AND SECURITY ENGINEERING POSSIBILITIES FOR EXPERIMENTAL TESTING OF ALARM TRANSMISSION SYSTEMS 123 M. Boros, M. Kucera , A. Velas, J. Valouch ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.3-14 3

Victor Aulin - Andrii Hrynkiv - Oleg Lyashuk - Yuriy Vovk - Sergii Lysenko - Dmytro Holub - Taras Zamota Andrii Pankov - Mariana Sokol - Vadym Ratynskyi - Olena Lavrentieva INCREASING THE FUNCTIONING EFFICIENCY OF THE WORKING WAREHOUSE OF THE "UVK UKRAINE" COMPANY TRANSPORT AND LOGISTICS CENTER

Competition in the field of logistics and individual transport services requires the constant analysis of logistics chains and their individual links. Given analysis enables implementation of new approaches to logistics companies and networks trade marketing in the offtake of certain types of goods. Industrial and commercial enterprises have to expand their trade and economic relations with other regions with the aim to provide the necessary sales volumes in the domestic market of regions and to go beyond it. Integration of the country's economy and the free trade policy orientation greatly simplify the task of establishing the free trade and economic relations between producers and consumers. However, there is a question of how quickly and efficiently; with minimal logistics costs, the goods at the right region of the customer should be delivered. The experiment on a one-storey closed warehouse No 3 of Brovary "UVK Ukraine", where a process of the cross- docking warehousing, with subsorting from a warehouse, has been conducted. The warehouse mainly works on trading networks. The loading of the warehouse has been performed with the input flow intensity within (12.3 ... 23.56 (t/h)). During the experiment, the number gain of loading and unloading machines (Electrostacks Reath Track OMG NEOS LAT 3.0) a pcs in the process of unloading, loading and moving consignments, has been examined. In addition, the increasing number of workers employed in non-mechanized forms of work, such as receiving and stacking cargo in the reception area (moving the relevant consignments to designated areas of the reception area and its arrangement), which in total is an increase of 4 employees in the technological process of the warehouse operation, has been investigated during the experiment. The growing of labor and mechanized resources makes it possible to better organize the work of the warehouse, reduce total costs and increase cargo output flow. Conducting the process of warehouse work optimization on the formed technology and corresponding labor and mechanized resources caused an increase of its efficiency. It has been found that the optimal values of intensity of the incoming cargo flow will be 17.93 (t/h); the number of employees engaged in non-mechanized labor - 12 workers, and loading and unloading machines - 3 (pcs). Under such conditions, the warehouse functioning according to the desirability level will reach value 0.792. In this case, the cost value of warehouse work is 181.07 US$/h and cargo output flow intensity is 10.23 (t/h). Results of experimental studies have shown that implementation of the proposed method of increasing the efficiency of warehouse work, based on the company "UVK Ukraine" may, on average, achieve a total cost reduction of 17.2% and an increase in output flow by 34.6% and the input flow by 10.3%. Keywords: warehouse logistics, transport and logistics center, labor resources, loading and unloading machines, warehouse, cargo flow, costs, optimization

1 Introduction countries, the logistics sector in its broadest sense is ranked as the third among the economic sectors. Underestimating The current state of the logistics market development the role of logistics as a factor of influence on the efficiency of any country, including Ukraine, depends on the level of and competitiveness of the economy, worsens Ukraine’s its economy. The volume of services provided in the field position in international rankings. Thus, Ukraine in 2019 of logistics directly depends on the activity level of their in the World Bank’s Logistics Performance Index (LPI) is customers, the dynamics of production, domestic and ranked 67 in the world, [1]. One should consider that the foreign trade. For example, in the European Union (EU) logistic component’s share in the gross domestic product

Victor Aulin1, Andrii Hrynkiv1, Oleg Lyashuk2, Yuriy Vovk2,*, Sergii Lysenko1, Dmytro Holub1, Taras Zamota1, Andrii Pankov1, Mariana Sokol3, Vadym Ratynskyi4, Olena Lavrentieva5 1Department of Maintenance and Repair of Machines, Central Ukrainian National Technical University, Kropyvnytskyi, Ukraine 2Department of Automobiles, Ternopil Ivan Puluj National Technical University, Ukraine 3Foreign Languages Department, I. Ya. Gorbachevskyy Ternopil National Medical University, Ukraine 4Department of Accounting and Taxation, Ternopil Ivan Puluj National Technical University, Ukraine 5Department of General Technical Sciences and Vocational Training, Kryvyi Rih State Pedagogical University, Ukraine *E-mail of corresponding author: [email protected]

(GDP) is 11 - 16% in developed countries (in the EU and logistical and organizational problems associated with the United States - 13-17%, in China - 25%, in Japan - 7%). design and management of individual warehouses with As for Ukraine, most experts relate the economy of the different systems of their warehouse functions it is difficult country to the level of the “third world” countries, in which to describe the wide range of issues and various options for logistics costs can reach 41% of the GDP. This figure in solving them [10]. Ukraine ranges from 31 to 34%. At the same time, 69% of The warehouse logistics is a functionally important logistics costs accounted for transport (6.8 bln US$), 24% - stage in warehouse operations and it has a direct impact for warehousing (2.3 bln US$) and about 6% - for logistics on cost and performance in the supply chain [11-13]. flow management (0.6 bln US$) [1-2]. Therefore, one of the Due to inefficient operation of the warehouse complex, main areas of the modern transport policy is transition to freight and logistics companies may experience significant the transport logistics in the transport and logistics system inconvenience associated with the expectation between (TLS). TLS allows to provide comprehensive services arrival and shipment [14]. This, in turn, leads to unjustified to consumers of transport services, to create conditions costs for the company in the form of a delay fee (fine for for the combined transport development, to reduce the delayed transportation and additional costs for keeping environmental burden on the environment [3]. The TLS the truck driver in overtime) and delayed delivery of structure should consist of a set of interconnected elements subsequent orders [15]. In addition, long idling trucks in interaction at the regional and local levels. Regional queues is not environmentally sustainable as it increases the distribution centers, terminal complexes, transport and concentration of carbon dioxide [16-18], so it is advisable logistics centers (TLCs) are the main system elements of to organize transport and storage complexes with a transport infrastructure objects [4]. standardized loading and a warehouse work efficiency. The The relevance of organization and increase of efficiency main aim of transport and logistics centers is to minimize of the transport and logistics centers is significantly the cost of logistics centers operating and to reduce the growing. This is evidenced by existence and creation of cost of goods transportation [1, 19]. Implementation of the new large, modern companies that specialize in this type short-term strategies for realization of the cargo service of cargo delivery organization. They compete with each functions in the transport and logistics complex will reduce other, offering customers better, more economical, faster logistics costs and increase the capacity of the warehouse and better ways of processing orders, both domestically [20]. The modeling approach and proposed warehouse and internationally. This necessitates the development of scheduling rules are pre-developed and can be adopted by efficient warehouse technologies and the organization of any enterprise. warehouse operations and logistics management. Within the framework of forming an efficient TLS system, 3PL - TLC operators and cargo shippers have been considered and the quality of logistics services has 2 Literature review been evaluated. Thus, the current state of the market can be determined: 41% of logistics is in a state of stagnation; Taking into consideration current trends in development 30% - at the stage of formation and development; 18% - at of the logistics services, it can be stated that the TLC have the stage of formation; 1.8% - at the redistribution stage become the basic elements of the cargo management and [21]. Consequently, customers and logistics providers seek cargo flows [5]. In general, such centers play a coordinating to optimize costs, implement new technologies, improve and distributive role in cargo transportation. Introduction the quality of logistics services and establish effective of the new modern technologies necessitates creation communication between customers and logistics providers of a network of TLC infrastructural objects that perform to increase the loyalty level of the end consumers [22]. functions of interaction between types of transport and International companies in Ukraine suggest that imperfect organization of material distribution in the economic legislation - 74%; corruption factor - 51%; customs work - region [6]. They are the basic foundation of material flow 59%; poor quality of infrastructure - 62%; lack of professional management that provide interconnection with senders, staff - 24%; market monopolization - 14% are the main consumers, carriers, etc. obstacles in development of the TLS market. Ukrainian Major trends, such as timeliness, containerization companies also gave to the factors of imperfect legislation and e-commerce, lead to a steady increase in the volume and corruption factors the first place - 58 and 57%, then of goods that have to be transported in accordance with imperfect infrastructure - 49%, difficulties in customs regulated delivery schedules [7]. Supply chains in this case work - 45%, problems with staff experience - 26%, and are formed in such a way as to use fully adaptive storage monopolization - 17%. Among the factors that can greatly systems as one of the main components of the transportation influence the development of the TLS, experts point out to process [8]. The baggage handling at airports, mixed implementation of EU legal acts and regulations [1, 23-24]. with inclined conveyor trays in the distribution centers Analysis of the complex interactions that occur between of the postal industry, warehouses formation of retailers different components of the transport and logistics center is consignments through implementation of the e-commerce the main problem of warehouse work improving [25]. The is an example of the most successful implementations of basic condition is a uniform distribution associated with the such warehouse complexes [9]. Because of the diverse loading and unloading work required for the flow of goods

COMMUNICATIONS 2/2020 VOLUME 22 INCREASING THE FUNCTIONING EFFICIENCY OF THE WORKING WAREHOUSE OF THE... 5

within the logistics network. Its analysis is performed using of material flow management and reducing the cost of the Petri nets and sensitivity functions as the discrete event operating the TLC in TLN. A one-storey closed warehouse modeling and simulation structures [25-27]. The warehouse No 3 of Brovary “UVK Ukraine” has been chosen for the modeling work is aimed at identifying the innovative aspect experiment, where a process of cross-docking warehousing of using both graph theory and elements of sensitivity with subsorting from a warehouse, has been implemented. theory [28-31] and statistical optimization [32-33]. With The total area of the warehouse is 1080 m2 and its height help of mathematical tools to support the functional is 9 m. Type of storage: 4 tier shelve rack. The warehouse characteristics of warehouse systems, their strengths and mainly works on trading networks. This warehouse is part weaknesses have been identified. However, that aspect of the TLS where goods can be transported as by road makes it possible to implement individual storage units for transport and by rail through the container platform of the various technological processes. Therefore, it is desirable TLS. The railway and highway connections are brought to identify the critical factors in the local logistics chain to the container platform. From the container platform in the warehouse and to propose solutions to improve its the required containers have been sent to the warehouse efficiency [34]. Use of the warehouse database processing reception area and, if it is necessary the completed methods is well suited to set the basic characteristics of containers from the warehouse departure area have been individual warehouse units. At the same time, the main moved, to the TLC container platform and further transport logistic processes, involving both labor and machinery is already formed. should not be damaged [35-36]. It should be noted that The technological process of the studied warehouse direct modeling of cargo flows together with quantitative complex TLC functioning has the following sequence of analysis made it possible to identify delays in the logistics operations: chain and to determine indicators that will allow increasing 1) Arrival of cargo to the warehouse: preparation of the efficiency of the resource usage. The last one will technical means for receiving cargo, documents, increase the productivity of warehouse work [30, 37]. familiarization of warehouse workers with the With significant growth of interest in the logistics unloading plan; enterprises efficiency, the need for transport and logistics 2) Check of cargo packing integrity in the vehicle: centers arises even more, so adaptation and high efficiency fastenings and presence of appropriate seals and of their work for the diverse needs of logistics agents markings are checked, external damages are revealed; is a priority at present [25]. Nevertheless, this requires 3) Unloading of cargo using the loading and unloading efficient coordination of warehouse work and formation mechanisms (LUM); of the warehouse technological process with the minimum 4) Cargo reception and stacking in the reception area: required number of operations for human and mechanized moving the respective consignments of cargo to resources [32]. the designated places of the reception area and its Properly organized technological process of the TLC arrangement; operation should ensure: accurate and timely carrying 5) Cargo reception by quantity and quality: opening of out of quantitative and qualitative acceptance of goods; container, counting by quantity and verification of effective use of mechanization of loading-unloading and documentation, culling; transport-warehouse works; consistent and rhythmic 6) Selection and transfer of cargo to the storage and performance of warehouse operations, which contributes to completing area; the systematic loading of warehouse workers and creation 7) Storage of cargo in a warehouse in two modes (1-2.5 of favorable working conditions; rational warehousing of days, 2.5 and more); goods ensuring maximum use of warehouse volumes and 8) Completing and sending from the storage and reception areas; safety of goods; clear organization of centralized areas: a consignment is formed for the relevant order, delivery of goods. the quantity of the cargo and the type of container The aim of this study was to develop a methodology is determined, packaging and sealing is carried out, for formation of the warehouse work rational technology documents are being prepared; organization and warehouse work optimization. 9) Moving the finished cargo unit to the departure area; 10) Verification of cargo by quantity and compliance with documents; 3 Materials and methods of determining the 11) Loading of cargo into the vehicle and transfer of rational technology of service orders documents. Determination of the effective work organization is The transport and technological process is an important carried out by setting a value - total costs, which form a part of the transport and logistics network (TLN). It must plurality of values corresponding costs for each element of be implemented to successfully complete the entire range the process. of work and reduce the cost of operations in the TLC. In the process of active experiment influence of factors

In addition, it must be based on advanced technology has been investigated: X1 - the input flow intensity;2 X - and advanced methods. Solution to this problem involves the number of employees and X3 - the number of loading optimizing the production process for efficient organization and unloading machines in the warehouse instead of

VOLUME 22 COMMUNICATIONS 2/2020 6 AULIN et al.

Table 1 Formation of factors and their levels for experiment Levels Factors Lower (-1) Upper (+1)

X1 - cargo input flow intensity, t/h 12.3 23.56

X2 - the number of employees engaged in non-mechanized labor, empl. 8 12

X3 - the number of loading and unloading machines, pcs. 2 3 organizational and economic components of the warehouse qualitative description of the object of study. This coefficient activity. The testimonials or resulting features of experiment is calculated by the formula: are as follows: Y1 - total costs (US$) → min; Y2 - cargo 2 N N 2 vzl t 2 r 2 output flow intensity (t/h) → max. The set of factors and RY=-112 =-||ii--YY/ iiY , (4) vY e i ==1 ^ h i 1 ^ h o their levels are presented in Table 1. For the experiment, it has been decided to investigate where: 2 2 three factors and their two levels. The number of necessary vvzl, Y - dispersion of regression residues, response, experiments can be calculated by formula: respectively t YYii,,r Y - actual, average, estimated value of the response. n f Ne = 2 , (1) The standard error that characterizes the standard deviation of the studied regression coefficients from the where: mean is calculated by the formula:

2 - number of levels, N t 2 nf - number of factors. | YYii- i = 1 ^h1 According to Equation (1) it was determined that it Sb jr = N $ (5) 2 n - 2 is necessary to carry out 8 experiments for solving the XXij - r j | ^ h optimization problems. A plan for the full-scale experiment, i = 1 indicating the levels and factors, as well as the response where n - sample volume. functions of the experiment, has been formed. In order to From a statistical point of view the regression minimize the impact on response, experiments should be coefficients evaluation is carried out by Student’s criterion. performed in random order. This compares the calculated value to the table at a given The analysis of this experiment plan has been confidence level of 0.05 and the calculated average values performed using portable software (Statistica v.10.0.1011.0). of levels: The processing of the experimental results began with the b jr regression analysis, i.e. the model has been built and the taa,/f = 2 t 2,ftb , (6) Sb jr unknown coefficients of the regression equation have been determined: where:

b jr - estimated regression coefficients,

2 2 2 a - probability 0.95, Yb= 01++++bX$$$1 bX22 bX33 ^^^hhh. (2) f - average value of level. ++bX412$$X bX53$$1 Xb+ 52$$XX3 With a significant regression coefficient, the Student’s test is greater than a tabular one. It has been determined that effects of the interaction The calculation of the marginal error of deviation has of factors, based on their nature and content, are virtually been established from the following calculations: absent and therefore, they have not been included in

r the overall appearance of the model (2). The unknown T jf= tSa, $tb $ b jr . (7) constant coefficients of the regression equation have been determined using the least squares method. The obtained The interval for each regression coefficient has been model coefficients can be obtained by the following determined according to inequality: formulas:

rrrr r N N bbjj-+TT##jjb j . (8) bY0 ==||/;NbjiYN/; i ==11i N (3) Mathematical model fit to the experimental data, that 2 bYji2 = XNji /;f is, its adequacy has been defined by Fisher’s criteria F. | ^h i = 1 The calculation criterion should be greater than the table Description of factors and response using the criterion: mathematical model (2) is characterized by a coefficient 2 2 vX R f2 of determination, which should be at least 0.95, for a F ==2 2 $ $ Fa,ft$ bl , (9) vY 1 - R f1

COMMUNICATIONS 2/2020 VOLUME 22 INCREASING THE FUNCTIONING EFFICIENCY OF THE WORKING WAREHOUSE OF THE... 7

where: of cargo flows movement through the warehouse complex

N has been used. These operations are necessary to obtain 2 r 2 vX =-| XXi /f1 - dispersion factor, reliable research data on changes in the cost of improving e i = 1 ^ho warehouse operations, namely the total cost ($) Y1 → min;

f1 - degree average value, the output flow intensity (t/h) Y2 → max. For the experiment, the Brovary “UVK Ukraine”” warehouse No 3

N (for household goods) has been loaded in accordance with 2 r 2 vY =-| YYi /N --f1 1 - dispersion value, the data shown in Equation (12) and the experiment plan e i = 1 ^ho below (one experiment one working day of the warehouse) N - experiments number, during the period from 04/15/2019 to 04/22/2019. The value R2 - determination coefficient. functions have been calculated as averages over the entire To solve the optimization problem, the desirability experimental warehouse operating day. Increasing the function approach, introduced by E. Harrington, has been number of employees engaged in non-mechanized labor used [38]. This method has useful properties of continuity, has been used on operations: cargo reception and stacking monotony and smoothness. The method converts specific in the reception area (moving the respective consignments parameters to abstract numeric values. A logical function of cargo to the designated places of the reception area and as the basis for the calculation has been used: its arrangement) by changing the number of employees from 1 to 2; cargo reception by quantity and quality: dYii=-expexp - , (10) (opening of container, counting by quantity and verification ^^hh of documentation, culling) by changing the number of where: Y - function value. employees from 1 to 2; completing and sending from the Function (10) is characterized by the two saturation storage and reception areas (a consignment is formed portions (d → 0 and d → 1) and a linear portion for the relevant order, the quantity of the cargo and the (d → 0.2 and d → 0.63). For a better abstract representation type of container is determined, packaging and sealing of a function, it is necessary to break it into ranges where is carried out, documents are being prepared) increase the specific values of the scale correspond to the studied from 1 person to 3 employees. Increases in loading and indicators. To form multivariate optimization, the function unloading machines have been made by increasing the is formed by the following equation: number of Reath Track OMG NEOS LAT 3.0 electric

n stackers on unloading, loading and moving consignments n Zd= % i . (11) from 2 pieces to 3. The full factorial experiment makes i = 1 it possible for multiple measurements that meet certain conditions set optimal factors values. The number of Further study of this function should determine analysis measurements is 8 and the values of the factors are of the investigated values and factors of the process. The combined in all the possible variants. Advantages of a full value of the curve optimum condition is set outside the factorial experiment are the simplicity of the solution of curve level, within these limits, the most relevant factor the task and the statistical redundancy of the number of values are selected. The function method is visual. measurements, which reduces the influence of errors of To determine the objective function values, a individual measurements on the estimation of unknown warehouse restrictions system based on the experimental regression coefficients. and organizational data with the relevant factors values have been studied. Thus, they have been determined during the statistics data analysis of the “UVK Ukraine” warehouse 4 Results complex functioning. 4.1 Study results of the proposed approach ]Z12..31##X 23 56; to organization rational efficiency increase ] ]82##X 12; of the warehouse work ] [2 ##X33; (12) ] ]Y1 " min; Based on the obtained and processed experimental ] ]Y2 " max . data, a database has been formed (Table 2) to identify \ changes in the optimality criteria for the organization of a The research planning experiment of cargo flow at warehouse complex. The minimum and maximum values in Brovary “UVK Ukraine” warehouse complex has been used the peak period of “UVK Ukraine” company have been used to find the optimal conditions for the work organization. to form the flow of goods. The following factors have been selected: the input flow The experimental data processing using application intensity (t/h) X1, the number of employees engaged in software makes it possible to automate the calculations manual labor (work.) X2, number of loading and unloading according to the formulas given. Regression analysis of machines of Reath Track OMG NEOS LAT 3.0 stackers (pcs.) the experimental results is shown in Tables 3 and 4,

X3. For this purpose, the experimental data normalization insignificant coefficients in the tables were not recorded.

VOLUME 22 COMMUNICATIONS 2/2020 8 AULIN et al.

Table 2 Value of the input parameters of the experiment according to the basic variant during the peak load movement in the warehouse complex The levels of variation Cargo output flow Total costs, US$ Cargo input flow The number of The number of loading and intensity, t/h intensity, t/h employees, empl. unloading machines, pcs. Experiment X1 X3 X4 Y1 Y2 1 12.3 8 2 193.70 3.6 2 23.56 8 2 162.49 6.3 3 12.3 12 2 173.31 9.6 4 23.56 12 2 140.29 8.4 5 12.3 8 3 343.82 3.7 6 23.56 8 3 273.05 7.7 7 12.3 12 3 294.34 11.9 8 23.56 12 3 221.85 12.05

Table 3 The Regression analysis of experimental results for the optimizing trait Y1 R2 = 0.9988 - the determination coefficient of the regression model of the experimental data Regression Regression Standard Student’s Significance level p, Average interval Average interval coefficients coefficients error coefficient - 95% - 95% (value) (p<0.05)

-2 -3 b0 -23.80 38.86·10 61.25 10.39·10 -28.74 -18.87 -2 -4 b1 25.51 1.43·10 1787.49 3.56·10 25.33 25.69 -3 -3 b2 3.71 33.82·10 109.75 5.8·10 3.28 4.14 -3 -3 b3 15.46 135.27·10 114.31 5.56·10 13.74 17.18 -2 -4 -4 -2 -2 b4 -64.49·10 9.99·10 645.44 9.86·10 -65.76·10 -63.22·10 -2 -3 -4 -2 -2 b5 -350.93·10 3.99·10 878.11 7.25·10 356.01·10 -345.85·10 -2 -3 -3 -2 -2 b6 -44.13·10 11.25·10 39.22 16.23·10 -58.42·10 -29.83·10

Table 4 The Regression analysis of experimental results for the optimizing trait Y2 R2 = 0.9988 - the determination coefficient of the regression model of the experimental data

Regression Regression Standard Student’s Significance level p, Average interval Average interval coefficients coefficients error coefficient - 95% - 95% (value) (p<0.05)

-2 -3 b0 -21.584 21.59·10 99.97 6.37·10 -24.33 -18.84 -2 -3 -3 -2 -2 b1 -62.28·10 7.93·10 78.56 8.1·10 -72.35·10 -52.20·10 -2 -3 -3 -2 -2 b2 282.67·10 18.79·10 150.46 4.23·10 258.8·10 306.55·10 -2 -3 -3 -2 -2 b3 899.01·10 75.15·10 119.63 5.32·10 803.53·10 994.5·10 -2 -4 -3 -2 -2 b4 4.94·10 5.55·10 89.00 7.15·10 4.23·10 5.65·10 -2 -3 -3 -2 -2 b5 11.77·10 2.22·10 53.00 12.01·10 8.95·10 14.59·10 -2 -3 -3 -2 -2 b6 -96.87·10 6.25·10 155.00 4.11·10 -104.82·10 -88.93·10

For a more accurate reflection of the experimental composition at studied factors. In the general case, the results, the mathematical regression model has been regression equations have the form: complicated to the second order. The factors did not 2 2 interact, if the coefficient of determination was not lower YX11=-23..80 ++25 51 $$37. 1 X2 + 2 ^^-hh2 than 0.95. While analyzing data in Tables 3 and 4, it is ++15..46 $$XX3 64 49 10 $$12X - (13) ^ --h 2 2 possible to conclude that all the included factors are --350..93 $$10 XX13 44 13 $$10 XX23$ statistically significant, as evidenced by the level of their --2 2 2 significance. Therefore, the values of the coefficients can YX2 =-21..584 -+62 28 $$10 1 282.67 $$10 2 --2 2 ^ h 2 be included in the model that describes the process of $$XX2 ++899..01 10 $$3 49410 $$XX12+ (14) ^^hh--2 2 +-11..77 $$10 XX13$$96 87 10 $$XX23

COMMUNICATIONS 2/2020 VOLUME 22 INCREASING THE FUNCTIONING EFFICIENCY OF THE WORKING WAREHOUSE OF THE... 9

Table 5 The dispersion analysis of the experimental data for the optimizing trait Y1 Coefficients Dispersion Fisher’s criterion Significance level p, (p<0.05)

3 6 -5 b1 2.68·10 26.49·10 12.4·10 3 5 -4 b2 2.56·10 25.34·10 4.0·10 3 5 -4 b3 5.38·10 53.15·10 2.76·10 2 4 -4 b4 4.21·10 41.66·10 9.86·10 2 4 -4 b5 7.81·10 77.10·10 7.25·10 3 -2 b6 1.56 1.53·10 1.62·10

Table 6 The dispersion analysis of the experimental data for the optimizing trait Y2 Coefficients Dispersion Fisher’s criterion Significance level p, (p<0.05)

3 2 -8 b1 11.25·10 1.09·10 1.42·10 3 2 -15 b2 92.02·10 8.98·10 1.42·10 3 2 -8 b3 10.44·10 1.01·10 2.41.42·10 3 2 -12 b4 30.50·10 2.97·10 9.17·10 3 2 -12 b5 30.75·10 3.00·10 8.62·10 3 2 -14 b6 71.44·10 6.97·10 1.27·10

a) b)

c) d) Figure 1 Graphical display of the projected and observed balances and estimation of their magnitude: a) and b) - for the total cost function; c) and d) - for the output flow intensity

To assess the adequacy of these models, a variance describe the studied process of the warehouse functioning, analysis of the obtained experimental data has been so the significance level p, for each factor is below the performed and Fisher’s criterion has been determined. allowable level of 0.05. Realization of the dispersion analysis is reflected in Tables Visually estimate residual dispersion between the 5 and 6. obtained and estimated values by regression Equations (13) From Tables 5 and 6 it can be seen that the factors and (14). It should be noted that significant dispersion of included in the mathematical model (13), (14) adequately 10% of the maximum predicted values should not be present

VOLUME 22 COMMUNICATIONS 2/2020 10 AULIN et al.

a) b) c)

Figure 2 Graphic representation of surfaces value on the scale of desirability of the studied warehouse work: a) dependence of Z on X1 and X2; b) dependence of Z on X1 and X3; c) dependence of Z on X2 and X3

Поверхность и контур желательн. ; Метод: Сплайн-подгонка Поверхность и контур желательн. ; Метод: Сплайн-подгонка 12. 5 12. 5 Поверхность и контур желательн. ; Метод: Сплайн-подгонка 12. 0 12. 0 12. 5 11. 5 11. 5 12. 0 11. 0 11. 0 11. 5 10. 5 10. 5 > 0.8 11. 0 > 0.8

X2 10. 0 X2 10. 0 < 0. 725 < 0. 725 10. 5 > 0.8 9. 5 9. 5 < 0. 625 < 0. 625

X2 10. 0 < 0. 725 9. 0 9. 0 9. 5 < 0. 525 < 0. 525 < 0. 625 8.5 8.5 9. 0 < 0. 425 < 0. 425 < 0. 525 8. 0 8. 0 8.5 < 0. 325 < 0. 325 < 0. 425 7. 5 8. 0 7. 5 < 0. 325 10 12 14 16 18 20 22 24 26 10 12 14 16 18 20 22 24 26 7. 5 10 12 14 16 18 20 22 24 26 X1 X1 X1 3. 2 3. 2 3. 2 3. 2 3. 2 3. 2 3. 0 3. 0 3. 0 3. 0 3. 0 3. 0 2. 8 2. 8 2. 8 2. 8 2. 8 2. 8 2. 6 > 0.8 2. 6 > 0.8 2. 6 > 0.8 2. 6 > 0.8 2. 6 > 0.8 2. 6 > 0.8 X3 X3 X3 X3 < 0. 725 < 0.7 < 0. 725 < 0.7 X3 < 0. 725 X3 < 0.7 2. 4 2. 4 2. 4 2. 4 2. 4 2. 4 < 0. 625 < 0.5 < 0. 625 < 0.5 < 0. 625 < 0.5 2. 2 < 0. 525 2. 2 < 0.3 2. 2 < 0. 525 2. 2 < 0.3 2. 2 < 0. 525 2. 2 < 0.3 < 0. 425 < 0.1 2. 0 < 0. 425 2. 0 < 0.1 2. 0 < 0. 425 2. 0 < 0.1 2. 0 2. 0 < 0. 325 < 0 < 0. 325 < 0 < 0. 325 < 0 1. 8 1. 8 1. 8 1. 8 1. 8 1. 8 10 12 14 16 18 20 22 24 26 10 12 14 16 18 20 22 24 26 7.5 8.0 8.5 9.0 9.5 10 12 14 16 18 20 22 24 26 7.5 8.0 8.5 9.0 9.5 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 10.0 10.5 11.0 11.5 12.0 12.5 X1 10.0 10.5 11.0 11.5 12.0 12.5 X1 X1 X2 X2 X2 a) b) c)

Figure 3 Graphical representation of value levels on the scale of the investigated tribological supplement: a) dependence of Z

on X1 and X2; b) dependence of Z on X1 and X2; c) dependence of Z on X2 and X3

Figure 4 Graphical representation of the procedure for finding the optimal factors values of the warehouse work of the function profile - Z

COMMUNICATIONS 2/2020 VOLUME 22 INCREASING THE FUNCTIONING EFFICIENCY OF THE WORKING WAREHOUSE OF THE... 11

Figure 5 Average results of the warehouse functioning according to the base variant (a) and optimized (b): 1 - total costs (US$/h); 2 - the input flow intensity (t/h); 3 - the output flow intensity (t/h); 4 - number of employees engaged in non-mechanized labor (empl.); 5 - number of loading and unloading machines (un)

in the analyzed experiment data. A graphical representation optimality of the warehouse functioning and the positive of this is presented in Figure 1. solution of the applied problem. The results, which give an It can be seen that there are no significant data opportunity to evaluate the effectiveness of the warehouse dispersion, and therefore the mathematical model can be functioning, have been shown in Figure 5. considered as certain. The formed optimal variant (Figure 5) makes it possible The surface values of the investigated functions of to increase the efficiency of the warehouse functioning. the warehouse functioning process resultant features with There is a decrease in total costs and increase in the reflections of factors values, on the scale of desirability, are intensities of the input flow and the output flow. identical in character and are presented in Figure 2. Levels of the function values on the scale of desirability are shown in Figure 3. 5 Discussion Visually analyzing these graphs, it is possible to state unequivocally that the optimal parameters of the 5.1 Research results discussion of the rational warehouse functioning have been presented in the studied technology service orders formation ranges of factors values X1, X2, X3. To determine the optimal values of the warehouse work efficiency, set the An increase in efficiency of the warehouse is possible limits of the investigated values, which will include the due to the proper organization of its work and providing entire experimental database that is available for analysis. technological operations with appropriate resources. The Under these conditions, it is possible to find the required transposition of the warehouse into the cross-docking maximum values for the function. Consider the above for process with subsorting from the warehouse has been each factor and each function. Realization of the procedure considered. The warehouse loading has been performed for determining the optimum value of storage conditions is with an input flow intensity of 12.3...23.56 (t/h). Under presented in Figure 4. those conditions, the number of loading and unloading From Figure 4 can be seen that the optimal variant machines (Electrostacks Reath Track OMG NEOS LAT of the factors values is at the intersection of function 3.0) a pcs., in the process of unloading, loading and maximum value in the defined interval of each factor. With moving consignments, has been increased. The number such values of the studied factors, it is possible to state the of workers employed in non-mechanized forms of work

VOLUME 22 COMMUNICATIONS 2/2020 12 AULIN et al.

has been gained: cargo reception and stacking in the of the input flow intensity can be increased by 10.3%. The reception area (moving the respective consignments of given procedure of an increase in the efficiency of the cargo to the designated places of the reception area and warehouse work functioning on realization of the cross- its arrangement) by changing the number of employees docking procedure, with subsorting from a warehouse, from 1 to 2; cargo reception by quantity and quality: reflects the positive effect for the enterprise. (opening of container, counting by quantity and verification of documentation, culling) by changing the number of employees from 1 to 2; completing and sending from the 6 Conclusions storage and reception areas (a consignment is formed for the relevant order, the quantity of the cargo and the type of A. It has been determined that optimum values for container is determined, packaging and sealing is carried warehouse input flow intensity is 17.93 (t/h), the out, documents are being prepared) increase from 1 person number of employees engaged in the non-mechanized to 3 employees. The increase of its efficiency has been labor - 12 empl., loading and unloading machines 3 pcs, shown by conducting the process of the warehouse work so under such conditions the value of the warehouse optimization on the formed technology and corresponding work will be 181.07 US$/h, cargo output flow intensity labor and mechanized resources. After conducting the - 10.23 (t/h). experiment and the corresponding calculations, the optimal B. During the experiment on the warehouse optimization, values of the studied parameters have been formed. The an increase in efficiency of the warehouse operation cargo output flow intensity is within the variation range on the implementation of the cross-docking procedure, (12.3...23.56 (t/h)) the optimum value will be 17.93 (t/h); the with warehouse subsorting, has been recorded. It number of employees engaged in the non-mechanized labor reflects the positive effect for the enterprise, namely are within the range (8...12) - optimum values of 12 workers, reduction of the total costs by 17.2%, increase of the and loading and unloading machines (2...3) - optimal value output flow rate by 34.6%, and average input flow 3 (units), which will rationally ensure the technological intensity values of 10.3%. process of warehouse operation. Under such conditions, C. The experiment of an increase in the warehouse the average level of the warehouse operation will be 0.792. work efficiency, based on the division of labor and The optimal value of the warehouse work costs will be mechanized resources, makes it possible to organize 181.07 US$/h, and cargo output flow intensity will be 10.23 and adapt the warehouse work to the requirements of (t/h). Compared to the basic variant, optimized one allows trade networks and customers of transport logistics reducing the total costs by 17.2%, and the output flow systems, in more detail. intensity to increase by 34.6%, while the average values

References

[1] AULIN, V., LYASHUK, O., PAVLENKO, O., VELYKODNYI, D., HRYNKIV, A., LYSENKO, S., HOLUB, D., VOVK, Y., DZYURA, V., SOKOL, M. Realization of the logistic approach in the international cargo delivery system. Communications - Scientific Letters of the University of Zilina [online]. 2019, 21(2), p. 3-12. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/1462 [2] KARPENKO, O., HORBENKO, A., VOVK, Y., TSON, O. Research of the structure and trends in the development of the logistics market in Ukraine. Journal of Sustainable Development of Transport and Logistics [online]. 2017, 2(2), p. 57-66. ISSN 2520-2979. Available from: http://dx.doi.org/10.14254/jsdtl.2017.2-2.5 [3] SAVIN, G., BRONNIKOV, S. Design to the city transport and logistics system in the conditions of increase of rates of institutional and technological changes. Business Logistics in Modern Management. 2018, Proceedings of the 18th International Scientific Conference Business Logistics in Modern Management, p. 485-500. ISSN 1849-5931. [4] BAKHTIZIN, R., EVTUSHENKO, E., BURENINA, I., GAISINA, L., SAGITOV, S. Methodical approach to design of system of the logistic centers and wholesale warehouses at the regional level. Journal of Advanced Research in Law and Economics [online]. 2016, 7(1), p. 16-25. ISSN 2068-696X. Available from: https://journals.aserspublishing.eu/jarle/ article/view/235 [5] RUSHTON, A., CROUCHER, P., BAKER, P. The handbook of logistics and distribution management: understanding the supply chain. 5. ed. India: Replika Press Pvt Ltd, Kogan Page Publishers, 2014. ISBN 978-0749466275. [6] HESSE, M., RODRIGUE, J.-P.: The transport geography of logistics and freight distribution. Journal of Transport Geography [online]. 2004, 12(3), p. 171-184. ISSN 0966-6923, eISSN 1873-1236. Available from: https://doi.org/10.1016/j.jtrangeo.2003.12.004 [7] PIECYK, M. I., MCKINNON, A. C. Forecasting the carbon footprint of road freight transport in 2020. International Journal of Production Economics [online]. 2010, 128(1), p. 31-42. ISSN 0925-5273, eISSN 1873-7579. Available from: https://doi.org/10.1016/j.ijpe.2009.08.027

COMMUNICATIONS 2/2020 VOLUME 22 INCREASING THE FUNCTIONING EFFICIENCY OF THE WORKING WAREHOUSE OF THE... 13

[8] SIAU, K., TIAN, Y. Supply chains integration: architecture and enabling technologies. Journal of Computer Information Systems. 2004, 44(3), p. 67-72. ISSN 0887-4417. [9] DENNIS, W. T. Parcel and small package delivery industry. North Charleston, SC: CreateSpace Independent Publishing Platform, 2011. ISBN 978-1461021544. [10] KAPSKI, D., KASYANIK, V., LOBASHOV, O., VOLYNETS, A., KAPTSEVICH, O., GALKIN, A. Estimating the parameters of traffic flows on the basis of processing of localization data on the movement of vehicles. Communications - Scientific Letters of the University of Zilina [online]. 2019, 21(2), p. 89-99. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/1474 [11] NARASIMHAN, R., KIM, S. W. Information system utilization strategy for supply chain integration. Journal of Business Logistics [online]. 2001, 22(2), p. 51-75. eISSN 2158-1592. Available from: https://doi.org/10.1002/j.2158-1592.2001.tb00003.x [12] VELYCHKO, O., VELYCHKO, L. Logistical modelling of managerial decisions in social and marketing business systems. Journal of International Studies [online]. 2017, 10(3), p. 206-219. ISSN 0910-5476. Available from: https://doi.org/10.14254/2071-8330.2017/10-3/15 [13] KOT, S., GOLDBACH, I. R., SLUSARCZYK, B. Supply chain management in SMEs - Polish and Romanian approach. Economics and Sociology [online]. 2018, 11(4), p. 142-156. ISSN 2071-789X. Available from: https://doi.org/10.14254/2071-789X.2018/11-4/9 [14] LEVINSON, M. The Box: how the shipping container made the world smaller and the world economy bigger-with a new chapter by the author. 2. ed. USA: Princeton University Press, 2016. ISBN 978-0691136400. [15] WALISZEWSKI, J. M. Towards understanding the impacts of congestion pricing on urban trucking. PhD Thesis. Cambridge, MA, USA: Massachusetts Institute of Technology, 2005. [16] AULIN, V., HRYNKIV, A., LYSENKO, S., DYKHA, A., ZAMOTA, T., DZYURA, V. Exploring a possibility to control the stressed-strained state of cylinder liners in diesel engines by the tribotechnology of alignment. Eastern-European Journal of Enterprise Technologies [online]. 2019, 3(12-99), p. 6-16. ISSN 1729-3774, eISSN 1729-4061. Available from: https://doi.org/10.15587/1729-4061.2019.171619 [17] AULIN, V., HRYNKIV, A., LYSENKO, S., ROHOVSKII, I., CHERNOVOL, M., LYASHUK, O., ZAMOTA, T. Studying truck transmission oils using the method of thermaloxidative stability during vehicle operation. Eastern-European Journal of Enterprise Technologies [online]. 2019, 1(6-97), p. 6-12. ISSN 1729-3774, eISSN 1729-4061. Available from: https://doi.org/10.15587/1729-4061.2019.156150 [18] AULIN, V., HRINKIV, A., DYKHA, A., CHERNOVOL, M., LYASHUK, O., LYSENKO, S. Substantiation of diagnostic parameters for determining the technical condition of transmission assemblies in trucks. Eastern-European Journal of Enterprise Technologies [online]. 2018, 2(1-92), p. 4-13. ISSN 1729-3774, eISSN 1729-4061. Available from: https://doi.org/10.15587/1729-4061.2018.125349 [19] GRABARA, J., BAJDOR, P., OKWIET, B. Center as a tool for logistics activities in support of the company on the example of SME’s enterprise X. Polish Journal of Management Studies. 2010, 2, p. 35-43. ISSN 2081-7452. [20] WEIDINGER, F., BOYSEN, N., SCHNEIDER, M. Picker routing in the mixed-shelves warehouses of e-commerce retailers. European Journal of Operational Research [online]. 2019, 274(2), p. 501-515. ISSN 0377-2217. Available from: https://doi.org/10.1016/j.ejor.2018.10.021 [21] TANAKA, R., ISHIGAKI, A., SUZUKI, T., HAMADA, M., KAWAI, W. Shipping plan for apparel products using shipping record and just-in-time inventory at a logistics warehouse. In: 7th International Congress on Advanced Applied Informatics IIAI-AAI 2018 : proceedings. 2019. eISBN 9781538674475, p. 682-687. [22] GRAWE, S. J. Logistics innovation: a literature-based conceptual framework. The International Journal of Logistics Management [online]. 2009, 20(3), p. 360-377. ISSN 0957-4093. Available from: https://doi.org/10.1108/09574090911002823 [23] SMITH, D., SRINIVAS, S. A simulation-based evaluation of warehouse check-in strategies for improving inbound logistics operations. Simulation Modelling Practice and Theory [online]. 2019, 94, p. 303-320, 2019. ISSN 1569-190X, eISSN 1878-1462. Available from: https://doi.org/10.1016/j.simpat.2019.03.004 [24] KOVACS, G., KOT, S. Economic and social effects of novel supply chain concepts and virtual enterprises. Journal of International Studies [online]. 2017, 10(1), p. 237-254. ISSN 2071-8330, eISSN 2306-3483. Available from: https://doi.org/10.14254/2071-8330.2017/10-1/17 [25] GERINI, C., SCIOMACHEN, A. Evaluation of the flow of goods at a warehouse logistic department by Petri Nets. Flexible Services and Manufacturing Journal [online]. 2019, 31(2), p. 354-380. ISSN 1936-6582, eISSN 1936-6590. Available from: https://doi.org/10.1007/s10696-018-9312-3 [26] MOODY, J. O., ANTSAKLIS, P. J. Supervisory control of discrete event systems using Petri nets [online]. New York: Springer Science & Business Media, 2012. ISBN 978-0-7923-8199-0, eISBN 978-1-4615-5711-1. Available from: https://doi.org/10.1007/978-1-4615-5711-1 [27] AULIN, V., LYASHUK, O., TYKHYI, A., KARPUSHYN, S., DENYSIUK, N. Influence of rheological properties of a soil layer adjacent to the working body cutting element on the mechanism of soil cultivation. Acta Technologica Agriculturae [online]. 2018, 21(4), p. 153-159. ISSN 1338-5267. Available from: https://doi.org/10.2478/ata-2018-0028

VOLUME 22 COMMUNICATIONS 2/2020 14 AULIN et al.

[28] OH, S. Comparative study on value determining factors of factory and logistics warehouse. International Journal of Innovative Technology and Exploring Engineering. 2019, 8(8S2), p. 1025-1030. ISSN 2278-3075. [29] PETKOVIC, T., PULJIZ, D., MARKOVIC, I., HEIN, B. Human intention estimation based on hidden Markov model motion validation for safe flexible robotized warehouses.Robotics and Computer-Integrated Manufacturing [online]. 2019, 57, p. 182-196. ISSN 0736-5845, eISSN 1879-2537. Available from: https://doi.org/10.1016/j.rcim.2018.11.004 [30] AULIN, V., LYSENKO, S., LYASHUK, O., HRINKIV, A., VELYKODNYI, D., VOVK, Y., HOLUB, D., CHERNAI, A. Wear resistance increase of samples tribomating in oil composite with geo modifier KGMF-1.Tribology in Industry [online]. 2019, 41(2), p. 156-165. ISSN 0354-8996, eISSN 2217-7965. Available from: https://doi.org/10.24874/ti.2019.41.02.02 [31] CAMPOLONGO, F., BRADDOCK, R. The use of graph theory in the sensitivity analysis of the model output: a second order screening method. Reliability Engineering & System Safety. 1999, 64(1), p. 1-12. ISSN 0951-8320. [32] BOYSEN, N., BRISKORN, D., FEDTKE, S., SCHMICKERATH, M. Automated sortation conveyors: A survey from an operational research perspective. European Journal of Operational Research [online]. 2019, 276(3), p. 796-815. ISSN 0377-2217. Available from: https://doi.org/10.1016/j.ejor.2018.08.014 [33] GLOVER, F., KLINGMAN, D., PHILLIPS, N. V. Network models in optimization and their applications in practice. 1. ed. USA: John Wiley & Sons, 1992. ISBN 978-0471571384. [34] BOYSEN, N., de KOSTER, R., WEIDINGER, F. Warehousing in the e-commerce era: a survey. European Journal of Operational Research [online]. 2019, 277(2), p. 396-411. ISSN 0377-2217. Available from: https://doi.org/10.1016/j.ejor.2018.08.023 [35] MAHROOF, K. A human-centric perspective exploring the readiness towards smart warehousing: the case of a large retail distribution warehouse. International Journal of Information Management [online]. 2019, 45, p. 176-190. ISSN 0268-4012, eISNN 0143-6236. Available from: https://doi.org/10.1016/j.ijinfomgt.2018.11.008 [36] YOSHITAKE, H., KAMOSHIDA, R., NAGASHIMA, Y. New automated guided vehicle system using real-time holonic scheduling for warehouse picking. IEEE Robotics and Automation Letters [online]. 2019, 4(2), 8618437, p. 1045-1052. ISSN 377-3766. Available from: https://doi.org/10.1109/lra.2019.2894001 [37] HARMON, R. L. Reinventing the warehouse: world class distribution logistics. 1. ed. The Free Press, 1993. ISBN 978-0029138632. [38] HARRINGTON, Jr., E. C. The desirability function. Industrial Quality Control. 1965, 21, p. 494-498. ISSN 0884-822X.

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.15-22 15

Silvia Dutkova - Karol Achimsky - Pawel Drozdziel SIMULATION OF A QUEUING SYSTEM OF A POST OFFICE IN ANYLOGIC SOFTWARE

This paper displays the design and application of a model that simulates the queuing system of a fictional post office. Starting point for solving more complicated optimization tasks is to create a system model that consists of elements of reality and the relationships between these elements. The key part of the paper includes the model of a queuing system of a post office created in Anylogic simulation software. The model of the post office displays post office with 5 postal counters, a certain input flow and a certain service time with an exponential probability distribution. The model also includes statistics and cost calculation. Keywords: post office, simulation, Anylogic, costs, conceptual model

1 Introduction the requirements is that a customer should not wait more than 12 minutes at the queue in the post office. Waiting and Digital technologies and to them related processes of service times at post offices are directly related to customer automation and digitalization, are the driving force of the satisfaction [1, 12]. present. They affect the labour market, productivity and Customer satisfaction, which co-creates a good image efficiency growth and ultimately are one of the key elements of the postal company and can significantly influence of transformation of functioning of society, businesses and customer behaviour and preferences, is one of the factors economies. Digitalization and automation also affect the of the of the postal services quality [22]. It is therefore transport sector which includes the postal companies [1]. necessary to shift focus to this area in order to retain There are many applications in the field of digitalization customers and meet their needs. and automation. From automated sorting systems and The other factor of the postal services quality is intelligent handling equipment in the processing of delivery location of the facility (post office). The success of each items to innovative delivery methods through autonomous provider of high-quality services depends on the location of vehicles and smart mailboxes in the process of mail delivery facilities of these businesses in relation to other facilities [2]. In the process of collection of delivery items and within and its customers. The choice of the facility is primarily the points of the first contact, the queuing systems are based on the availability, with customers choosing the commonly used. It is possible to design such systems closest (available) facility. Given that the availability through use of the computer simulation. Objectives of provides an overall indicator of how a particular facility is computer simulation are: elimination of the shortcomings accessible for other sources located in the analysed space, of analytical methods and capturing of properties and it is also necessary to determine the boundaries of this elements of the real system that cannot be expressed in the indicator [23]. analytical solution. In the system such as queuing system of a post office there are many random variables that cannot be captured in the analytical solution of optimization 2 Theoretical background problems [3-9]. Therefore it is advisable to use a simulation method to get the system model closer to the real system, 2.1 Queuing theory as much as possible. Contribution of the simulation method is in the ability of capturing the dynamic side of the The queuing theory is a mathematical discipline that system and complicated probabilistic relationships [10-15]. analyses and solves processes in which request flows In Slovakia, the quality of postal services is monitored and flow through certain devices from which they require the quality requirements for the universal postal service service. The pioneer of queuing theory was Danish engineer are regulated by the regulatory authority of the Slovak Agner K. Erlang. At the beginning of the 20th century, Republic [16-21]. Many of those quality requirements affect he derived unique patterns for probabilities of states the Queuing systems of post offices. For example, one of of a stabilized Markov system with losses. The queuing

Silvia Dutkova1,*, Karol Achimsky1, Pawel Drozdziel2 1Faculty of Operation and Economics of Transport and Communications, University of Zilina, Slovakia 2Lublin University of Technology, Poland *E-mail of corresponding author: [email protected]

Figure 1 General queuing system systems are systems (physical, social) serving to meet the 3 Data and methods needs of individuals, customers and requirements entering the system. The queuing service system is everything that The fundamental methods that are used in this paper is between the arrival of a request in the system and its are modelling and simulation method. In addition to the departure from the system. A queuing system of a post simulation model in the Anylogic program, the paper office can be considered as a stochastic system with displays a conceptual model of the queuing system of a post theoretically unlimited queue, a certain average service office. The conceptual model of the system was created time, a customer arrival and a certain number of service before the simulation model itself. lines [1, 7-9]. The model in Anylogic software displays five postal The queuing theory deals with problems, which involve counters servicing customers. Three counters are universal queuing (or waiting) in the line. Typical examples might be: and provide to customers all the postal services from the • banks/supermarkets -waiting for service, scope of universal service. There is also a financial counter • computers - waiting for a response, and a service counter at the post office. For the public the • public transport - waiting for a train or a bus. opening hours (every work day) are from 7:00 to 19:00. Modelling of queuing systems requires introduction The queuing system of a fictional post office is of notions originated by Kendall, which are also used to characterized by the two main random variables, namely describe a queuing system [24-25]: the average time of service and the average customer input at the post office. For this paper the average customer input A/ B / m / K / n/ D, (1) is given:

where: A - distribution function of the inter arrival λ = 2.20 minutes. (2) times, B - distribution function of the service times, where m The given average service time is: is a number of servers, K - capacity of the system, the maximum number of µ = 4.20 minutes. (3) customers in the system including the one being serviced, where n is the population size, which determines the Input data of the model input are fictional. The number of sources of customers, simulation of a particular system requires a detailed system D - service discipline. analysis that includes both an analysis of customer input Exponentially distributed random variables are flow and service time. expressed by M, meaning Markovain or memoryless [22-23]. According to the Kendall’s classification of queuing Generally, all the queuing systems can be divided systems, the queuing system of a post office is classified as into individual sub-systems consisting of entities, which a system with an infinite queue, five independent parallel are queuing for some activity (see Figure 1) [24-25]. lines, exponential distribution of the arrivals of customers, service times and a waiting discipline FIFO.

2.2 System modelling 3.1 Conceptual model of the post office queuing Simulation methods are methods that are often used system in the queuing theory. Solving complicated optimization tasks using simulation often leads to a creation of model The model starts with an event - the arrival of the that contains elements of a real system and relationships customer. The customer checks whether the counters are between them. The dynamic elements of the system are occupied. If they are not occupied, the customer enters handled by the time-slicing method. This method can counter. If they are currently occupied, he/she chooses model the progress of time when an event occurs. An event a minimum queue. It is clear that the customer chooses represents a change in the state of the system at a time. a counter according to the service he/she requires. If the There are two event types in simpler event-based simulation customer is waiting in the queue for more than 6.5 minutes, models arrival of customer and an end of service [10-13]. he/she leaves the post office without being served. This

COMMUNICATIONS 2/2020 VOLUME 22 SIMULATION OF A QUEUING SYSTEM OF A POST OFFICE IN ANYLOGIC SOFTWARE 17

Figure 2 Conceptual model of the post office queuing system condition is based on the results of a survey conducted in employee works with an information system. With this 2018 among postal customers. If a given counter is free, system, employee is able to provide service to the customer. the customer enters the counter to be served. A post office After the customer is served, he or she pays for the service.

VOLUME 22 COMMUNICATIONS 2/2020 18 DUTKOVA et al.

Figure 3 The queuing model compiled in Anylogic simulation software

Figure 4 Calculation of income and expenses in software Anylogic

The customer is then served and the model ends with 5.1 Simulation model customer’s activity - customer is leaving the post office. The whole process of functioning of the described model is The simulation model of a queuing system of a fictional presented in Figure 2. post office, designed in Anylogic is graphically interpreted in Figure 3. The simulation model (Figure 3) consists of following 4 Objective objects: Customer_Input - which is defined by interval time. In The main objective of this paper is to design an this case it is exponential distribution with one parameter. application of a model simulating the queuing system of The average customer input is λ = 2.20 minutes. Customer a fictional post office. Using the program, five simulation comes into the system (queuing model) for service. If they experiments were conducted. Based on the results of wait in the queue for more than 6.5 minutes, they leave simulation experiments, it is possible to make further the system without being served (refusal of service). This decisions about the system. Decisions are derived from means the loss for the queuing system of the queuing model. the simulation experiments of a simulation model, which Decision 1 - customer has to decide if he wants includes several variables and graphs. to use service type number 1 provided by Service_line and Service_line2 or service type number 2 provided by Service_line3 and Service_line4. Probability for service 5 Results type 1 is set up at 0.60. Probability of service type 2 is set up at 0.40. Design of the simulation model and results of five Queue - after decision 1, customer chooses the queue. simulation experiments are presented in this section. The The logic is in always choosing the minimum queue. If structure of the simulation model corresponds to structure queues are 0, the customer chooses the service line which of the conceptual model, which was defined in Figure 2. is free.

COMMUNICATIONS 2/2020 VOLUME 22 SIMULATION OF A QUEUING SYSTEM OF A POST OFFICE IN ANYLOGIC SOFTWARE 19

Table 1 Input data for simulation experiment 1 Time/minutes Simulation Time 3600 Service Time 4.2 Customer Input 2.2

Table 2 Input data for simulation experiment 2 Time/minutes Simulation Time 3600 Service Time 4.2 Customer Input 4.4

Figure 5 Simulation 1 -utilization statistics

Service lines - are defined by the delay time which • The final number of customers arriving at the post is in this case the average service time with exponential office within five business days is 7076 customers, distribution. In this case the average service time is µ = 4.20 • The first three service lines were able to serve 4283 minutes. Customers enter the counter providing the service customers. According to customer numbers and from the queue (unless they decide to leave because of individual service lines one can see that 74% of a long waiting time). customers wanted the first type of services, which Finance - object Finance 5 is only for calculating were served by Service_line and only 3.5% of customers expenses and revenue. Figure 4 displays calculation of were served by Service_line2, income and expenses. The input data are fictional. • The similar situation happened in the case of Service_ Model contains exits 1-4. There is also exit 5 for line3 and Service_line4. Only 2.69% of customers wanted customers that decide to leave the queue after 6.5 minutes. the second type of services, which were served by Service_line4. Figure 5 displays percentage utilization of Service_line3 and Service_line4, respectively queue 5.2 Simulation results 3 and queue 4.

Several simulation experiments were performed in Anylogic simulation software. The first simulation 5.2.2 Simulation experiment 2 experiment is based on input data from section 3. The rest of the simulation experiments are based on “what if” What if the customer input would be doubled? Solution questions and are solved in the following parts of paper. to this problem provides the simulation experiment 2. Table 2 displays input data of the simulation experiment 2. In 5.2.1 Simulation experiment 1 order to solve the problem the customer input is doubled. Simulation experiment 1 reflects current situation The simulation experiment 2 provided following in terms of queueing in the service lines in a fictional statistics about the system: post office. Table 1 display input data of the simulation • The final number of customers arriving in the post experiment 1. Simulation time (3600 minutes) represents office within 5 business days is 14 625, five business days. Opening hours in working days are 12 • The first three service lines were able to serve 8744 hours a day. customers. Only 8.5% customers were served by the The simulation experiment 1 provided following Service_line2. It can be said that doubled customer statistics about the system: input doesn´t have huge effect on the Service_line2,

VOLUME 22 COMMUNICATIONS 2/2020 20 DUTKOVA et al.

Table 3 Input data for simulation experiment 3 Time/minutes Simulation Time 3600 Service Time 4.2 Customer Input 6.6

Table 4 Input data for simulation experiment 4 Time/minutes Simulation Time 3600 Service Time 4.2 Customer Input 2.2

Figure 6 Average waiting time statistics in simulation experiment 4

• regarding theService_line4, the percentage change is to this problem. The Table 4 displays input data of the even less, only 1.6%. simulation experiment 4. 5.2.3 Simulation experiment 3 The simulation experiment 4 provided following statistics about the system: Simulation experiment 3 solves the following: what if • the waiting time for the service type 1 is no longer the customer input would be tripled because of Christmas? than 2.23 minutes even when Service_line2 is closed. Table 3 displays input data of the simulation experiment 3. System was able to serve these customers without The customer input data is tripled. any losses. Figure 6 displays statistics related to The simulation experiment 3 provided following waiting time, for example the average waiting time or statistics about the system: maximum or minimum waiting time in queue. • only 14.5% of customers were served by the Service_ line2. It can be said that tripled customer input doesn´t have huge effect on the Service_line2, 5.2.5 Simulation experiment 5 • utilization of the Service_line2 doesn´t seem efficient; the same case is in the Service_line4, only 18% of Last problem solved in this paper is bound to the customers were serviced by the Service_line4. question: What if the Service_line4 is closed? The input data of simulation experiment 5 are the same as in simulation experiment 4. 5.2.4 Simulation experiment 4 The simulation experiment 5 provided following statistics about the system: Another problem appears: What if Service_line 2 is • the waiting time for service type 2 is no longer than closed? The simulation experiment 4 provides the solution 3.10 minutes even when Service_line4 is closed. The

COMMUNICATIONS 2/2020 VOLUME 22 SIMULATION OF A QUEUING SYSTEM OF A POST OFFICE IN ANYLOGIC SOFTWARE 21

system was able to serve these customers without any system will be still able to serve customers at a customer losses. friendly waiting time. The major results obtained by the system provide the results that are helpful for making more 6 Conclusion and discussion decisions in the future. The main objective of this paper was to design the The paper examines the queuing system of fictional application of the model simulating the queuing system of post office. Anylogic simulation software has proven to a fictional post office. The simulation model is functional be a useful tool for modelling and simulating a queuing and provides several statistics. This model also serves system. Using the program, five simulation experiments as a background for further research oriented to specific were conducted. These experiments were based on “what post office with specific input data. This kind of research if” questions. “What if” questions were oriented to a change requires a detailed analysis of a specific queuing system, in customer input flow or a change in the number of postal identification of input parameters, their probability counters. In both cases of a change in the customer input distribution, etc. An important phase in modelling the flow, the change does not have a huge effect on the system. system is to build a model that matches the queuing system The postal counters were able to serve doubled and tripled and creating it in simulation software. In the future, it number of customers without any losses. The service_line2 is advisable to use Anylogic software for such research. productivity was low and only 3.5% of customers wanting Subject-oriented simulation experiments can result in service type 1 were served by this postal counter. Closing optimization of the queuing system, new compilation of Service_line2 was done without any losses. Same situation work schedules for post office employees, change in the can be seen in the case of Service_line4. Closing Service_ number of postal counters or it can increase the system line4 was done without any losses. The productivity of productivity. system can be increased by closing Service_lines2 and 4 and

References

[1] CHINORACKY, R., COREJOVA, T. Impact of digital technologies on labor market and the transport. Transportation Research Procedia[online]. 2019, 40, p. 994-1001. ISSN 2352-1465.Available from: https://doi.org/10.1016/j.trpro.2019.07.139 [2] TURSKA, S., MADLENAKOVA, L. 2019. Concept of smart postal mailbox. Transportation Research Procedia [online].2019, 40, p. 1199-1207. ISSN 2352-1465.Available from: https://doi.org/10.1016/j.trpro.2019.07.167 [3] DUTKOVA S., ACHIMSKY, K., HOSTAKOVA, D. Simulation of queuing system of post office. Transportation Research Procedia [online]. 2019, 40, p. 1037-1044.ISSN 2352-1465. Available from: https://doi.org/10.1016/j.trpro.2019.07.145 [4] DUTKOVA, S., HOSTAKOVA, D., MISOK, T., RYBICKA, I. Determination of probability distribution of customer input at post office. Transport and Communications. 2017, 5(2), p. 6-10. ISSN 1339-5130. [5] ACHIMSKY, K. Simulacne modely vo vyuke a ich vyuzitie pri rieseni problemov prepravy posty a PNS / Simulation models in education and their use in solving problems of post office transport (in Slovak). In: Celostatny vedecky seminar F-PEDaS VSDS / National Scientific Seminar FPEDaS VSDS : proceedings. 1988. p. 115-117. [6] ACHIMSKY, K. Prirucka pre uzivatelov / User Guide (in Slovak). In: Optimalizacia oblastnej postovej prepravnej siete / Optimization of the regional post transport network. Zilina: VU KS VSDS, 1985. [7] KRPAN, L., MARSANIC, R., MILKOVIC, M.A model of the dimensioning of the number of service places at parking lot entrances by using the Queuing theory. Technicki Vjesnik / Technical Gazette [online]. 2017, 24(1), p. 231-238. ISSN 1330-3651, eISSN 1848-6339. Available from: https://doi.org/10.17559/TV-20160128161848 [8] ACHIMSKY, K. Simulacia cinnosti postovej priehradky / Simulation of postal counter (in Slovak). In: Zbornik vydany na pocest zivotneho jubilea prof. RNDr. Michala Haranta/ Proceedings Issued in Honor of the Jubilee of Life prof. RNDr. Michal Harant : proceedings. l990. p. 127-133. [9] HU, X., BARNES, S., GOLDEN, B. Applying queuing theory to the study of emergency department operations: a survey and discussion of comparable simulation studies. International Transactions in Operational Research [online]. 2018, 25(1), p. 4-59. ISSN 0969-6016, eISSN 1475-3995. Available from: https://doi.org/10.1111/itor.12400 [10] BAHADORI, M, MOHAMMADNEJHAD, S. M., RAVANGARD, R., TEYMOURZADEH, E. Using Queuing theory and simulation model to optimize hospital pharmacy performance. Iranian Red Crescent Medical Journal [online]. 2014, 16(3), e16807. ISSN 2074-1804, eISSN 2074-1812. Available from: https://doi.org/10.5812/ircmj.16807 [11] KNESSL, C., VAN LEEUWAARDEN, J. S. H. Transient analysis of the Erland A model. Mathematical Methods of Operations Research [online]. 2015, 82, p. 143-173. ISSN 1432-2994, eISSN 1432-5217. Available from: https://doi.org/10.1007/s00186-015-0498-9 [12] HUSEK, R., LAUBER, J. Simulacny modely / Simulation models (in Czech). Praha: STNL/ALFA, 1987. ISBN07-326-87. [13] MADLENAK, R., MADLENAKOVA, L. Comparison of regional postal transportation networks in Zilina region. In: 19th International Scientific Conference on Transport Means : proceedings. 2015. p. 277-280.

VOLUME 22 COMMUNICATIONS 2/2020 22 DUTKOVA et al.

[14] BREZAVSCEK, A., BAGGIA, A. Optimization of a call centre performance using the stochastic queuing models. Business Systems Research [online]. 2014, 5(3), p. 6-18. ISSN 1847-9375. Available from: https://doi.org/10.2478/bsrj-2014-0016 [15] MARSANIC, R., ZENZEROVIC, Z., MRNJAVAC, E. Application of the Queuing theory in the planning of optimal number of servers (ramps) in closed parking systems. Economic Research / Ekonomska Istrazivanja [online]. 2015, 24(2), p. 26-43. ISSN 1331-677X, eISSN 1848-9664. Available from: https://doi.org/10.1080/1331677X.2011.11517453 [16] MADLENAK, R., STEFUNKO, J. The optimization approach of postal transportation network based on incapacitated fixed charge location model in conditions of Slovak republic. Transport Problems [online]. 2015, 10(4), p. 35-43. ISSN 1896-0596, eISSN 2300-861X. Available from: https://doi.org/10.21307/tp-2015-046 [17] MADLENAKOVA, L., TURSKA S., MADLENAK R. The image of the postal company as a key attribute of the customer’s purchasing behaviour. Transportation Research Procedia [online].2019, 40, p. 1088-1095. ISSN 2352-1465.Available from: https://doi.org/10.1016/j.trpro.2019.07.152 [18] DROZDZIEL, P., WINSKA, M., MADLENAK, R., SZUMSKI, P. Optimization of the post logistics network and location of the local distribution center in selected area of the Lublin province. Procedia Engineering [online]. 2017, 192, p. 130-135.ISSN 1877-7058. Available from: https://doi.org/10.1016/j.proeng.2017.06.023 [19] MADLENAK, R., MADLENAKOVA, L., STEFUNKO, J. The variant approach to the optimization of the postal transportation network in the conditions of the Slovak republic. Transport and Telecommunication Journal [online]. 2015, 16(3), p. 237-245. ISSN 1407-6179. Available from: https://doi.org/10.1515/ttj-2015-0022 [20] MADLENAK, R., MADLENAKOVA, L., STEFUNKO, J., KEIL, R. 2016. Multiple approaches of solving allocation problems on postal transportation network in conditions of large countries. Transport and Telecommunication Journal [online].2016, 17(3), p. 222-230.ISSN 1407-6179. Available from: https://doi.org/10.1515/ttj-2016-0020 [21] CHINORACKY, R., COREJOVA, T. Impact of digital technologies on labor market and the transport. Transportation Research Procedia [online]. 2019, 40, p. 994-1001. ISSN 2352-1465.Available from: https://doi.org/10.1016/j.trpro.2019.07.139 [22] TURSKA, S., MADLENAKOVA, L. 2019. Concept of smart postal mailbox. Transportation Research Procedia [online].2019, 40, p. 1199-1207. ISSN 2352-1465.Available from: https://doi.org/10.1016/j.trpro.2019.07.167 [23] HOSTAKOVA, D., MADLENAK, R., DUTKOVA, S. 2019 Indicator availability of postal services determined through gravity methods. Transportation Research Procedia [online]. 2019, 40, p. 244-250. ISSN 2352-1465.Available from: https://doi.org/10.1016/j.trpro.2019.07.037 [24] BOSE, S. K. An introduction to queueing systems. New York: Kluwer Academic/Plenum Publishers, 2002. ISBN 978-1-4615-0001-8. [25] BOLCH, G., GREINER, S., DE MEER, H., TRIVEDI, K. Queueing networks and Markov chains. 2. ed. New York: Wiley & Sons, 2006. ISBN 0-471-56525-3.

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.23-30 23

Kamil Krasuski - Stepan Savchuk ACCURACY ASSESSMENT OF AIRCRAFT POSITIONING USING THE DUAL-FREQUENCY GPS CODE OBSERVATIONS IN AVIATION

This study publishes results of tests with regard to determination of the aircraft positioning accuracy by means of the GPS navigation in aviation. The research exploits the mathematical model of the linear combination "Ionosphere- Free" in order to designate the coordinates of an aircraft. The research uses the actual GPS code observations, recorded by a satellite receiver mounted in the Cessna 172, at the time of the experiment for the EPDE military aerodrome in Dęblin. The computations of the position of the Cessna 172 aircraft for the linear combination "Ionosphere-Free" were made in the APS Toolbox v.1.0.0. programme. Within evaluation of accuracy of the GPS positioning in aviation, the determined coordinates of the aircraft Cessna 172 from the APS programme were compared to an accurate reference position from the solution derived by the PPP measurement technique. In the research, the authors obtained an average positioning accuracy of approximately 5 m in the geocentric XYZ coordinates and approximately 4 m in the ellipsoidal BLh coordinates. In addition, the 3D-error parameter is lower than 7 m for the XYZ geocentric coordinates. Keywords: GPS, Ionosphere-Free linear combination, accuracy, PPP method, aviation

1 Introduction plane, and 37 m for navigation in the vertical plane [5]. Furthermore, the time synchronization error in the GPS In the 21st century, the GNSS satellite technology system during a flight operation must not be worse than became a common method of aircraft positioning in the 40 ns (approximately 12 m). In addition, error of the GPS field of air navigation and air transport. Development of satellite’s position in the orbit must not be greater than modern GNSS global navigation systems, such as GPS, 30 m, whereas an error of determining the GPS satellite GLONASS, BeiDou, Galileo, QZSS Zenith and augmentation speed and acceleration must not exceed values of 0.006 systems SBAS (EGNOS, SDCM, WAAS, MSAS, GAGAN, m/s and 0.002 m/s2, respectively. In turn, availability of NAVIC) enabled their full operation and implementation the constellation of the GPS satellites during the flight in the process of determining the aircraft position. The operations should exceed 99%. The credibility parameter framework of application, operation and implementation of of the GPS system functioning in the civil aviation should the GNSS satellite technology in aviation has been clearly exceed 99.79% across the whole globe. The likelihood defined by the International Civil Aviation Organization of the lack of continuity of the navigation solution of an (ICAO). The Annex 10 to the Chicago Convention currently aircraft position by means of the GPS observation must allows the use of the GNSS satellite technique in aviation not exceed 2·10-4 per hour [6]. It should be mentioned that for the needs of performing air operations within the the criteria of the parameters of accuracy, availability, navigation systems of: GPS and GLONASS as GNSS satellite reliability and continuity are referenced to the standard systems which are certified by the ICAO and augmentation SPS positioning service in the GPS navigation system [7]. systems: ABAS, SBAS, GBAS [1-3]. Within the Standard Positioning Service, it is possible to use In the case of the global GNSS navigation systems, L1-C/A observation codes to determine the coordinates of only the GPS and GLONASS satellite systems are certified an aircraft in the GPS system. It must be noted that the GPS for a general use in civil aviation. It should be emphasized satellites transmit the signal L1-C/A on the carrier frequency that the GPS and GLONASS systems are fully operational of 1575.42 MHz, using the code-division multiple access and provide continuous satellite positioning for its users technique (CDMA) [8]. across the globe for 24 hours a day. Certification of the The main aim of this article is to determine the GPS and GLONASS navigation systems in civil aviation positioning accuracy of the aircraft Cessna 172 based includes such parameters as accuracy, availability, on the GPS solution. An analysis of the accuracy of the reliability and continuity [4]. In accordance with the ICAO aircraft positioning was carried out based on the actual recommendation, the accuracy of determining the position navigation data from the GPS satellite system, registered of an aircraft with the use of the GPS navigation system by an airborne receiver Topcon HiperPro, mounted in an must not exceed 17 m for navigation in the horizontal aircraft Cessna 172. The position of the Cessna 172 was

Kamil Krasuski*, Stepan Savchuk Institute of Navigation, Military University of Aviation, Deblin, Poland *E-mail of corresponding author: [email protected]

d11$ tgBi- determined based on the linear combination Ionosphere- z Ba=+rc tan 2 ; t 2 $ tg Bi-1 Free for dual-frequency GPS code observations. The e d o (3) y t numerical calculations were carried out in individually La==rc tan ; h - R, a x k cos B developed software called Aircraft Positioning Software (APS) Toolbox, developed in the language environment where: Scilab 6.0.0. (a,b)- semi-major and semi-minor axes of the ellipsoid frame, respectively, ab22- e - eccentricity, e = 2 , 2 Research method a R - radius of curvature of the prime vertical,

The fundamental observation equation of the linear a R = 22, combination Ionosphere-Free, which facilitates the 1 - eB$ sin 22 possibility of determining an aircraft position in the GPS t =+xy, satellite system, can be expressed as [9]: ae$ d1 = , t $ 1 - e2 PP31=+=aa122 Pd+-cd$ tr dts + ^h (1) 1 ++Trop RelM+ P3, d2 = , 1 - e2 where: i-1 - previous iteration, P - the “Ionosphere-Free” linear combination for the GPS 3 (B,L,h) - geodetic coordinates of aircraft’s position, code observations, B - Latitude, aa12, - linear coefficients, ^h st nd L - Longitude, (P1, P2) - code observations at the 1 and 2 frequency in the GPS system, h - ellipsoidal height. d - geometric distance between satellite and receiver, 222 dx=-XySS+-Yz+-ZS , ^^hh^ h (x,y,z) - coordinates of the aircraft in XYZ geocentric frame, 3 Research experiment

(XS,YS,ZS) - satellite coordinates in GPS system, c - light speed, The numerical tests for the presented research dtr - receiver clock bias correction, method were performed for the actual navigation and dts - satellite clock bias correction, GPS observation data, obtained from the airborne Topcon Trop- tropospheric delay, HiperPro receiver, mounted in an aircraft Cessna 172. Rel - relativistic effect, The test flight by the Cessna 172 was executed over the

MP3- multipath effect for the GPS code observations. military airport EPDE, in Dęblin. Duration of the flight was approximately one hour, with the maximum speed of Coordinates of the aircraft from Equation (1) are the flight reaching 80 m/s. The geodetic Topcon HiperPro determined using the least squares method in a stochastic receiver worked in the kinematic mode, gathering the GPS model, as [10]: data at a frequency of 1 s. In order to reduce the effect of Z multipath interference and radio disturbances, the Topcon ]AQ$ xd-=lv ] T HiperPro receiver was installed in the cockpit just behind ]NA= $$pA [ , (2) the plexiglass. The GPS code observations, stored in the ]L = ApT $$dl ] memory card of the geodetic receiver Topcon HiperPro, ]Qx = NL-1 $ \ were used to reproduce the position of the aircraft Cessna where: 172 in the post-processing mode. The numerical calculations

Qx- vector with unknown parameters, to determine the position of the Cessna 172 were made in N = AT . p . A - matrix of the normal equations frame, the authors’ original software APS (Aircraft Positioning A - matrix of coefficients, matrix is full rank, Software) Toolbox v.1.0.0, which was fully developed p - matrix of weights, in the Scilab 6.0.0. environment. For the purpose of the L = AT . p . dl - vector of absolute terms, conducted numerical calculations, the configuration of the dl - vector with difference between measurements and APS programme was set, as shown in Table 1. modeled parameters, v - vector of residuals. The determined aircraft coordinates from Equations 4 Results and discussion (1) and (2) are physically referenced to the geocentric XYZ frame. Alternatively, there is a possibility to express aircraft The determined Cessna 172 coordinates in APS coordinates in the ellipsoidal BLh frame. In this case, it is programme, in the geocentric XYZ frame and in the necessary to use a transformation between the geocentric ellipsoidal BLh frame, were compared to the actual reference XYZ frame and the ellipsoidal BLh frame in iterative trajectory of the aircraft flight in order to determine solution, as shown below [11]: the accuracy of the GPS signal in aviation. The precise

COMMUNICATIONS 2/2020 VOLUME 22 ACCURACY ASSESSMENT OF AIRCRAFT POSITIONING USING THE DUAL-FREQUENCY GPS CODE... 25

Table 1 Configuration of computations in the APS software Parameter Configuration GNSS system GPS system type of observations code observations at the 1st and 2nd frequency in GPS system type of RINEX file 2.10 source of satellite ephemeris data precise ephemeris from the IGS service source of satellite clock data precise ephemeris from the IGS service method of satellite position computation 9-degrees Lagrange polynomial method of satellite clock bias computation 9-degrees Lagrange polynomial effect of Earth rotation and time of pseudorange travelling through applied atmosphere relativistic effect applied troposphere source Simple model receiver clock bias estimated multipath and measurement noise not applied satellite and receiver phase center offset based on ANTEX file from IGS service Sagnac effect Applied Cut-off elevation 50 a priori standard deviation of pseudorange observations 1 m positioning mode kinematic mathematical model of solution least square estimation in iterative scheme adjustment processing Applied maximum number of iterations in a single measurement epoch N=10 number of unknown parameters k=4, for each measurement epoch number of observations n>4, for each measurement epoch interval of computations 1 s initial coordinates of aircraft position based on header of RINEX file time of the GPS system GPS Time reference frame IGS’08 format of output coordinates geocentric XYZ and ellipsoidal BLh local test of residuals applied global statistical test test Chi-square significance level 10-=a .95 ^h DOP coefficients estimated

coefficients value for HPL and VPL level kHPL = 6.18 and kVPL = 5.33 reference flight trajectory of the Cessna 172 was determined delay file format for the computations. The position of the using the PPP precision measurement technique. The PPP Cessna 172 in the CSRS-PPP programme is expressed in measurement technique uses the mathematical model of the geocentric XYZ, as well as ellipsoid BLh coordinates, in the linear combination equation - “Ionosphere-Free”. The the global frame IGS’08, whereas the scale of the reference computations of the reference position of the aircraft time is the GPS Time. It is worth adding that the program Cessna 172 for the PPP measurement technique were made CSRS-PPP allows an accurate determination of kinematic in the CSRS-PPP programme [12]. In the framework of the positioning for the aircraft with mean errors of coordinates calculations performed in the CSRS-PPP programme, the at a level of approximately 10 cm [13]. authors used dual frequency P1/P2 code and L1/L2 phase Accuracy of the GPS code positioning for the presented observations from the airborne receiver Topcon HiperPro. research method in the APS programme was defined The CSRS-PPP programme uses the IGS precise products both in the geocentric XYZ coordinate frame and in the such as SP3 precise ephemeris, precise satellite clocks ellipsoidal BLh frame. The accuracy of the GPS positioning CLK, the characteristics of the antenna phase center of the in geocentric XYZ coordinates is specified as [14]: satellite/receiver ANTEX file format, and DCB hardware

VOLUME 22 COMMUNICATIONS 2/2020 26 KRASUSKI, SAVCHUK

Figure 1 Accuracy of the Cessna 172 aircraft positioning in the XYZ geocentric coordinates

Z ]dx =-xxPPP the RMS error along the axis Z is equal to 1.9 m. The span of ] [dy =-yyPPP , (4) the GPS positioning accuracy along the X axis is over 17.5 ] ]dz =-zzPPP m, along the Y axis it is approximately 6 m, and along the \ Z axis it is over 15 m. It is worth noting that approximately where: 75% of the accuracy of GPS positioning results along the X (dx,dy,dz) - accuracy of the GPS positioning in the XYZ axis are less than 0. On the other hand, approximately 97% geocentric coordinate frame, results of the GPS positioning accuracy along the axis Y is x - coordinate of the aircraft along the X axis, determined less than 0. Moreover, approximately 99% results of the GPS based on Equation (1), positioning accuracy along the axis Z is less than 0. It can y - coordinate of the aircraft along the Y axis, determined therefore be concluded that results obtained with regard to based on Equation (1), accuracy of the GPS positioning in the APS programme are z - coordinate of the aircraft along the Z axis, determined lower than the readings of the aircraft Cessna 172 position based on Equation (1), in the PPP solutions in the CSRS-PPP programme. xPPP - coordinate of the aircraft along the X axis, determined Figure 2 shows another parameter of accuracy of the from the PPP measurement technique in the CSRS-PPP GPS positioning in the geocentric XYZ frame, in the form programme, of the 3D-error. The 3D-error parameter specifies a shift in yPPP - coordinate of the aircraft along the Y axis, determined coordinates of the aircraft Cessna 172 between a solution from the PPP measurement technique in the CSRS-PPP in the APS programme and in the CSRS-PPP programme programme, in the geocentric XYZ frame. The mathematical formula zPPP - coordinate of the aircraft along the Z axis, determined specifying the value of the 3D-error parameter can be from the PPP measurement technique in the CSRS-PPP determined as [15]: programme. Figure 1 shows results of the GPS positioning 3De-=rrordxd22++ydz2 . (5) accuracy in the geocentric XYZ coordinates for the linear combination “Ionosphere-Free”, based on a solution in Based on the obtained results, it is possible to observe the APS programme. The positioning accuracy along the X that the value of the 3D-error parameter ranges from 1.2 axis in the GPS system for the receiver Topcon HiperPro m to 16.1 m. In addition, the average value of the 3D-error ranges from - 11.8 m to + 5.7 m. In addition, the value of the parameter equals 6.9 m. It is worth stressing that the average positioning accuracy along the X axis in the GPS value of the 3D-error parameter towards the end of the system equals - 2.4 m. Moreover, the RMS error along the X observation time decreases to a level below 10 m. The drop axis is equal to 3.4 m. The positioning accuracy along the Y in the numerical value of the 3D-error parameter is relevant axis for the Topcon HiperPro receiver ranges from - 5.2 m since it directly translates into the location of the aircraft in to + 0.9 m. In addition, the value of the average positioning the three-dimensional space. accuracy along the Y axis in the GPS system is - 2.3 m. Figure 3 shows results of the GPS positioning Additionally, the RMS error along the Y axis is equal to accuracy in the ellipsoidal BLh coordinates for the linear 1.1 m. The positioning accuracy along the axis Z for the combination “Ionosphere-Free” based on a solution in the Topcon HiperPro receiver ranges from - 13.3 m to + 2.0 m. APS programme. Accuracy of the GPS positioning in the Moreover, the value of the average positioning accuracy ellipsoidal BLh coordinates is specified as [16]: along the Z axis in the GPS system is - 5.1 m. Furthermore,

COMMUNICATIONS 2/2020 VOLUME 22 ACCURACY ASSESSMENT OF AIRCRAFT POSITIONING USING THE DUAL-FREQUENCY GPS CODE... 27

Figure 2 Results of 3D-error parameter

Figure 3 Accuracy of the Cessna 172 aircraft positioning in the BLh ellipsoidal coordinates

Z ]dBB=-BPPP The positioning accuracy for the B component in the ] [dLL=-LPPP , (6) GPS system for the receiver Topcon HiperPro ranges from ] ]dhh=-hPPP - 6.8 m to + 5.4 m. In addition, the average value of the \ where: positioning accuracy along the axis B in the GPS navigation (dB,dL,dh) - accuracy of the GPS positioning in the system is + 0.6 m. Moreover, the RMS error along the ellipsoidal BLh frame, axis B is equal to 2.7 m. The positioning accuracy for the B - coordinate of the aircraft for the B component, component L, for the receiver Topcon HiperPro, ranges determined based on Equation (3), from - 4.2 m to + 2.2 m. Furthermore, the average value L - coordinate of the aircraft for the L component, of the positioning accuracy along the axis L in the GPS determined based on Equation (3), navigation system equals - 1.2 m. In addition, the RMS error h - coordinate of the aircraft for the h component, along the axis L is equal to 0.9 m. The positioning accuracy, determined based on Equation (3), for the component h for the receiver Topcon HiperPro,

BPPP - coordinate of the aircraft along the B axis, determined ranges from - 15.1 m to + 5.5 m. In addition, the average by the PPP measurement technique in the CSRS-PPP value of positioning accuracy along the axis h in the GPS programme, navigation system is - 4.1 m. Besides, the RMS error, along

LPPP - coordinate of the aircraft along the L axis, determined the axis h, is equal to 3.1 m. It is worth stressing that only by the PPP measurement technique in the CSRS-PPP 38% of the GPS positioning accuracy findings, along the axis programme, B, are below 0. Therefore, readings of the coordinate B in hPPP - coordinate of the aircraft along the h axis, determined the APS programme are inflated in relation to results in the by the PPP measurement technique in the CSRS-PPP programme CSRS-PPP. On the other hand, approximately programme. 90% results of the GPS positioning accuracy along the axis L

VOLUME 22 COMMUNICATIONS 2/2020 28 KRASUSKI, SAVCHUK

are less than 0. Moreover, approximately 95% results of the 5 Conclusions GPS positioning accuracy along the h axis are less than 0. It can therefore be concluded that the obtained findings with The article presents and describes results of the regard to accuracy of the GPS positioning for the L and h scientific research on use of the dual-frequency GPS components in the APS programme are too low in relation code observations in aircraft positioning. In particular, to the measurements of the aircraft Cessna 172 in the PPP the study focuses on determination of accuracy of the solution, in the CSRS-PPP programme. GPS positioning in air transport. During the research, the Based on the conducted investigations, it appears that authors used a linear combination “Ionosphere-Free” as the largest RMS error in the geocentric XYZ coordinates is a mathematical model to determine aircraft coordinates. along the X axis. On the other hand, the lowest RMS error The source materials for the scientific research used the in the geocentric XYZ coordinates is along the Y axis. In raw P1/P2 code observations in the GPS system from the the coordinate ellipsoidal BLh frame, the largest RMS error Topcon HiperPro receiver. The Topcon HiperPro geodetic can be observed for the coordinate h, while the lowest one receiver was installed on board the Cessna 172, which is along the axis L. The value of the RMS parameter results performed a test flight over the EPDE military airport in directly from the obtained results in the accuracy of GPS Dęblin. The collected code P1/P2 observations in the GPS positioning, in the geocentric XYZ frame or in the ellipsoidal system allowed performing numerical calculations in order BLh frame, respectively. In the case of air operations, it is to establish coordinates of the Cessna 172. The numerical essential to monitor changes of coordinates of an aircraft calculations of the Cessna 172 position were made in flight trajectory, on a continuous basis. Regardless ofthe the APS v.1.0.0 software programme. For the purpose of selected method (or technique) of the GPS positioning determining the GPS positioning accuracy in aviation, the in aviation, the key issue is credibility of the determined authors made a comparison of the determined coordinates aircraft coordinates. Thus, in aviation research conducting from the APS reference position of the Cessna 172. The an objective control of the navigation parameters recorded precise flight trajectory of the Cessna 172 was determined by sensors, during the execution of the flight operations, based on the PPP measurement technique in the CSRS-PPP plays a tremendous role. Mounting a GPS receiver on board programme. Based on computations, it was found that the an aircraft allows monitoring changes in the navigation accuracy of the GPS positioning in aviation is: position, in real time and post factum. For this reason, it • between - 13.3 m and + 5.7 m in the geocentric XYZ is important to develop navigation applications for control coordinates, calculations of an aircraft position, such as the APS • between - 15.1 m and + 5.5 m in the ellipsoidal BLh programme, from the beginning. coordinates. The presented research method was applied in papers In addition, the accuracy measurements of determining by one of this paper co-authors. For example, in paper the position in the form of the RMS parameter are as [17], the Iono-Free linear combination was utilized for follows: recovery of the aircraft position based on dual-frequency • less than 3.4 m in the geocentric XYZ coordinates, GPS code observation in the APS software. In that paper • less than 3.1 m in the ellipsoidal BLh coordinates. the parameters such as: aircraft trajectory, receiver clock The obtained research findings indicate a problem with bias, DOP coefficients, mean errors of aircraft ellipsoidal the low GPS positioning accuracy in aviation. The problem coordinates, Chi-square statistical test, HPL and VPL is closely related to the choice of a positioning method integrity terms were presented and described. In the presented in this work. The dual-frequency GPS code next paper [18], the Iono-Free linear combination was observations have high measurement noise at the carrier utilized for recovery of the aircraft position based on the frequency L1 and L2. Moreover, in the case of an application dual-frequency GPS code observation in gLAB software of the dual-frequency GPS code observations in navigation package. In that paper, the parameters such as: aircraft computations, there appears an issue of eliminating an trajectory, DOP coefficients, mean errors of aircraft ionospheric delay, which directly affects determination ellipsoidal coordinates, MRSE term, HPL and VPL integrity of the horizontal coordinates of an aircraft. The authors terms were analyzed and shown. In another article [19], the plans to conduct research into monitoring the ionospheric Iono-Free linear combination for the BSSD module in the status during the air operations in the airspace of Poland. APS software package was applied for determination of Moreover, in the future additional research is planned the aircraft position based on the GPS data. In that paper aimed at exploiting the GLONASS code observations in the parameters such as: DOP coefficients, mean errors of aircraft positioning in aviation. aircraft ellipsoidal and geocentric coordinates, MRSE term, HPL and VPL integrity terms, Chi-square statistical test were estimated and presented. Acknowledgement

This paper was supported by Military University of Aviation for 2020 year.

COMMUNICATIONS 2/2020 VOLUME 22 ACCURACY ASSESSMENT OF AIRCRAFT POSITIONING USING THE DUAL-FREQUENCY GPS CODE... 29

References

[1] FELLNER, A., JAFERNIK, H., TROMINSKI, P. RNAV GNSS essential for the LUN implementation and the chance for the Polish General Aviation (in Polish). Transactions of the Institute of Aviation. 2011, 211, p. 57-67. ISSN 0509-6669, eISSN 2300-5408. [2] JAFERNIK, H., KRASUSKI, K. and MICHTA, J. Assessment of suitability of radionavigation devices used in air (in Polish). Scientific Journal of Silesian University of Technology. Series Transport [online]. 2016, 90, p. 99-112. ISSN 0209-3324, eISSN 2450-1549. Available from: https://doi.org/10.20858/sjsutst.2016.90.9 [3] KRZYKOWSKA, K., SIERGIEJCZYK, M. Signal monitoring as a part of maintenance of navigation support system in civil aviation. Archives of Transport System Telematics. 2014, 7(1), p. 36-40. ISSN 1899-8208. [4] FELLNER, R. Analysis of the EGNOS/GNSS parameters in selected aspects of Polish transport. Transport Problems. 2014, 4(9), p. 27-37. ISSN 1896-0596 [5] JAFERNIK, H., FELLNER, A., MROZIK, M., KRASUSKI, K. Results of aircraft positioning tests in post-processing using GNSS. Scientific Journal of Silesian University of Technology. Series Transport [online]. 2017, 95, p. 67-74. ISSN 0209-3324, eISSN 2450-1549. Available from: https://doi.org/10.20858/sjsutst.2017.95.7 [6] ICAO Standards and Recommended Practices (SARPS). Annex 10, Volume I (Radionavigation aids) - International Civil Aviation Organization. 2006, Polish version. Available from: http://www.ulc.gov.pl/pl/prawo/ prawomi%C4%99dzynarodowe/206-konwencje, current on: 15.12.2016 [7] GRALL, P., SPECHT, C. Comparative evaluation of GPS SPS standards for accuracy in fixing position (in Polish). Zeszyty Naukowe Akademii Marynarki Wojennej. 2011, 3(186), p. 43-45. ISSN 2657-6260, eISSN 2657-7291. [8] JANUSZEWSKI, J. Compatibility and interoperability of Satellite Navigation Systems in different modes in transport (in Polish). Prace Naukowe Politechniki Warszawskiej. 2013, 95, p. 199-208. ISSN 1230-9265. [9] GAO, Y., CHEN, K. Performance analysis of precise point positioning using real-time orbit and clock products. Journal of Global Positioning System [online]. 2004, 3(1-2), p. 95-100. eISSN 1446-3164. Available from: https://doi.org/10.5081/ jgps.3.1.95 [10] SANZ SUBIRANA, J., JUAN ZORNOZA J. M., HERNANDEZ-PAJARES M. Fundamentals and algorithms. Vol. 1. In: GNSS data processing. Noordwijk, Netherlands: ESTEC, ESA Communications, 2013. ISBN 978-92-9221-886-7, p. 139- 144. [11] OSADA, E. Geodesy (in Polish). Wroclaw: Oficyna Wydawnicza Politechniki Wroclawskiej, 2001. ISBN 83-7085-663-2, p. 236-241. [12] MALINOWSKI, M., KWIECIEN, J. A comparative study of precise point positioning (PPP) accuracy using online services. Reports on Geodesy and Geoinformatics [online]. 2016, 102(1), p. 15-31. ISSN 2391-8152. Available from: http://dx.doi.org/10.1515/rgg-2016-0025 [13] KRASUSKI, K. Utilization CSRS-PPP software for recovery aircraft’s position. Scientific Journal of Silesian University of Technology. Series Transport [online]. 2015, 89, p. 61-68. ISSN 0209-3324, eISSN 2450-1549. Available from: https://doi.org/10.20858/sjsutst.2015.89.7 [14] BAKULA, M. Static code DGPS positioning based on three reference stations. Geodesy and Cartography. 2005, 54(2), p. 81-92. ISSN 1648-3502. [15] RODRIGUEZ-BILBAO, I., RADICELLA, S. M., RODRIGUEZ-CADEROT, G., HERRAIZ, M. Precise point positioning performance in the presence of the 28 October 2003 sudden increase in total electron content. Space Weather [online]. 2015, 13, p. 698-708. eISSN 1542-7390. Available from: https://doi.org/10.1002/2015SW001201 [16] SWIATEK, A., JAWORSKI, L. and TOMASIK, L. EGNOS Monitoring prepared in space research centre P.A.S. for SPMS project. Artificial Satellites [online]. 2017, 52(4), p. 109-120. ISSN 2083-6104. Available from: https://doi.org/10.1515/ arsa-2017-0010 [17] KRASUSKI, K. Application of PPP-KINEMATIC method and aircraft positioning software for calculation of plane’s position, Przeglad Techniczny Uzbrojenia. 2016, 140(4), p. 53-68. ISSN 1230-3801. [18] KRASUSKI, K., JAFERNIK, H. Implementation GNSS satellite technique in air transport – example (in Polish). In: Characteristics of selected research aspects in theory and practice. Katowice: Wydawnictwo Naukowe Sophia, 2017. ISBN 978-83-65357-91-5, p. 21-31. [19] CWIKLAK, J., GRZEGORZEWSKI, M., KRASUSKI, K. The application of the BSSD iono-free linear combination method in the processing of aircraft positioning, Journal of KONES Powertrain and Transport [online]. 2019, 26(3), p. 15-22. ISSN 2354-0133. Available from: https://doi.org/10.2478/kones-2019-0052

VOLUME 22 COMMUNICATIONS 2/2020 30 KRASUSKI, SAVCHUK

Annex - Nomenclature

The abbreviation The full name GLONASS Globalnaja Nawigacionnaja Sputnikowaja Sistiema / Global Navigation Satellite System GPS Global Positioning System BSSD Between Satellite Single Difference ICAO International Civil Aviation Organization GNSS Global Navigation Satellite System ABAS Aircraft Based Augmentation System SBAS Satellite Based Augmentation System GBAS Ground Based Augmentation System EPDE ICAO airport code APS Aircraft Positioning Software IGS International GNSS Service ANTEX Antenna Exchange Format RINEX Receiver Independent Exchange System HPL Horizontal Protection Level VPL Vertical Protection Level DOP Dilution of Precision MRSE Mean Radial Spherical Error PPP Precise Point Positioning XYZ Geocentric coordinates BLh Latitude, Longitude, ellipsoidal height BeiDou Chinese satellite navigation system Galileo European satellite navigation system QZSS Zenith Japanese regional satellite navigation system EGNOS European Geostationary Navigation Overlay Service SDCM The System for Differential Corrections and Monitoring WAAS The Wide Area Augmentation System MSAS The Multi-functional Satellite Augmentation System GAGAN The GPS Aided Geo Augmented Navigation NAVIC Indian Regional Navigational Satellite System SPS Standard Positioning Service C/A Coarse Acquisition CDMA Code-Division Multiple Acces CSRS-PPP Canadian Spatial Reference System - Precise Point Positioning DCB Differential Code Biases RMS Root Mean Square

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE hhttps://doi.org/10.26552/com.C.2020.2.31-41 31

Miroslava Mikusova - Jamshid Abdunazarov - Joanna Zukowska - Alisher Usmankulov DESIGNING OF PARKING SPACES TAKING INTO ACCOUNT THE PARAMETERS OF DESIGN VEHICLES IN RUSSIA

Nowadays, in all the cities, there is an acute problem of a lack of parking spaces. The number of vehicles are becoming more and more larger, not only in megacities, but in small cities of the country, as well and there are no more parking places - the pace of solving the problem is several times slower than the rate of the transport growth among the citizens. The article is dedicated to the determination of the optimum sizes parking place for designing vehicles on a parking space, which is an element of the roads. The optimum amount parking places are determined on examples of the passenger cars and trucks. The results of research on dimensioning of parking spaces and recommendations to use the results for design of objects of transportation infrastructure are presented. According to the research, authors included the term "design vehicle" and gave its definition. The authors developed a template for 7 types of design vehicles and their sizes and also recommended sizes for longitudinal parking for each of their design vehicles. The optimum parking plot angles are determined, as well. Keywords: trajectory, parking space, design vehicle, software Auto TURN, turning radius of vehicles

1 Introduction In the domestic regulatory and procedural documents, the dimensions of parking spaces for the road Trends in the size of cars in traffic flow and an acute infrastructure facilities are defined in the Methodological shortage of parking space require a more careful attitude Recommendations of the SRC MDRS MIA [3], IRM towards design of the size of a parking place and parking 218.4.005-20101 и SS Р 52289-20042. The dimensions in these space. Unfortunately, design of parking does not take documents were copied from the Handbook for Automobile into account the composition of the traffic flow that takes Transportation and Traffic Management [4] published in shape on a specific road, transport infrastructure object the USSR in 1981, which, in turn, was a translation of the (requirements are obvious here, in the USA where the size American Road Traffic Management Handbook of 1965 of cars is larger than in Europe, the size of parking space and the recommendations given in the third edition of the is larger), duration of parking is not taken into account; Transportation and Traffic Engineering Handbook [5]. short-term parking near shops, banks, etc., requires more Requirements for parking geometry in regulatory space for manoeuvring upon arrival and departure from the documents contain ambiguous, sometimes contradictory parking space than during the long-term parking). The most information that may adversely affect the level of the road acute problem manifested itself when a ban was imposed safety. Thus, in the “Methodological recommendations on on the transit movement of vehicles weighing more than the design and equipment of highways to ensure traffic 12 tons during the daytime along the Moscow Ring Road safety” [6], the turning radius of a passenger car is 8 m and (Resolution of the Mayor of Moscow, dated November for a truck is 9-12 m. When approximate calculation of the 15, 2012 No. 650-PP “On Amendments to Legal Acts of the total area of coverage in parking place, including the area Government of Moscow” [1]). According to the Moscow of manoeuvring and parking is done, it is recommended to mayor’s office, more than 150 thousand trucks with a proceed from the average area per one passenger car of maximum weight of more than 3.5 tons are moving through 25 m2 and on a truck 40 m2. At the same time, in the album the city streets during the daytime. About 40 thousand of typical projects “Cross-sectional profiles of highways trucks are arriving daily from other regions. passing through settlements” (TP503-0-47.86) [7], the At the Moscow Ring Road, large trucks make up 30% of average parking area for a truck should be 92.4 m2, not 40 the flow, half of which are transit and do not serve the needs m2, as stated in the methodological recommendations. The of the capital [2]. At that time, there was no experience dimensions of the parking space, given in the Regulations in designing parking place for cars arriving in Moscow or for the placement of multifunctional zones of road service following in transit. on roads [8-10], take into account the size of modern cars,

Miroslava Mikusova1,*, Jamshid Abdunazarov2, Joanna Zukowska3, Alisher Usmankulov4 1Department of Road and Urban Transport, Faculty of Operation and Economics of Transport and Communications, University of Zilina, Slovakia 2Jizzakh Polytechnic Institute, Faculty of Auto Transport, Department of Land Transport Systems, Uzbekistan 3Faculty of Civil and Environmental Engineering, Gdansk University of Technology, Poland 4Jizzakh Polytechnic Institute, Faculty of Industrial Technologies, Jizzakh, Uzbekistan *E-mail of corresponding author: [email protected]

Figure 1 Schemes of planning of parking spaces for passenger cars with one-sided (a) and two-sided (b) placement

Figure 2 The fundamental planning of placement of the parking spaces for trucks, recommended in the US: z - 30-45°; A - 25.9-3.0.5 m; B- 9.1-13.7m; C - 15.2-18.8 m; D - 9.1-13.7 m; E - 3.0.5-3.5.0 m. but this is not enough to develop a complete planning 30°, the width of the passages can be reduced to 6.0 m and solution, since the parking manoeuvres are not taken into the width of each parking space - by 30 cm. For large trucks, account, and only the dimensions of the parking space are the length of the longitudinal parking space must be at least given. 41 m and width 5.2 m. The same values are specified in the The situation in the regulatory documents of Uzbekistan regulations of the United Arab Emirates for large trucks on is similar. Since Uzbekistan was part of the USSR, it uses parking spaces. the same technical document that Russia used before, [3]. At the Department of Survey and design of roads MADI At the moment, technical documents, recommending the a research to justify the size of parking spaces for vehicles, size of design vehicles, taking into account the current taking into account the characteristics for modern traffic on flow in the republic, have not been adopted. For example, the roads of the Russian Federation, has been conducted. the parameters (sizes) of trucks that were produced in the The studies were conducted on category 1b roads, as well 70s of the last century are still being used. The situation as on city parking lots. is similar in terms of the size of the parking spaces. At This research work included: monitoring parking the moment, the republic uses the regulatory document, manoeuvres, studying the real situation when setting up which requires updating based on time. The passenger car parking spaces and modelling parking manoeuvres of and truck placement schemes, used in the United States passenger cars and trucks, using the Auto TURN software, and the dimensions of parking spaces shown in Figure 1, which allows simulating the movement and manoeuvring provide more complete information. The planning solution of vehicles at speeds up to 60 km/h, and also to model for placing the parking spaces for trucks, which provides the three-dimensional movement on a 3D surface, localize the simplest conditions for entering and leaving a parking modelling for various groups of vehicles; graphically space, recommended in the USA, is shown in Figure 2. The represent the dynamic dimensions indicating the dynamic sizes on the scheme, presented in Figure 3, correspond to dimensions of the vehicles (external and internal wheels, a parking angle of 45°, while it is indicated that at angles of characteristic points of the body); create vehicle reversal

COMMUNICATIONS 2/2020 VOLUME 22 DESIGNING OF PARKING SPACES TAKING INTO ACCOUNT THE PARAMETERS OF DESIGN VEHICLES... 33

Figure 3 Scheme of parking spaces for trucks patterns [11]. The AutoTURN is a CAD software released account the area of the exit and entry zones and the area of for the AutoCAD, MicroStation, BricsCAD platforms, the parking space itself. developed and marketed by Transoft Solutions Inc. Among Due to increase in the dynamic characteristics of many things, it analyzes and models the maneuvering of a vehicles, the requirements for the construction and design vehicle along a path. of roads are being improved. During the roads designing, it The parking space for vehicles includes parking space becomes necessary to introduce the concept of a “design for a vehicle and a manoeuvring area, designed for the vehicle”. This term is defined differently. “A design vehicle entrance to parking spaces, exit and setting cars. The is a vehicle used to determine the geometrical parameters dimensions of the parking space must ensure unhindered of roads (minimum turning radii of the intersection at one entry, opening the doors of vehicles, unloading or loading level, turning radii, roundabout) affecting safety, capacity luggage, and then unimpeded exit without hitting other and cost of the intersection. This is a conditional transport vehicles. unit, the parameters of which are used in the calculations The dimensions of the car parking space determine its of the pavement and its elements. A design vehicle is type and size (length, width, turning radius of the inner rear such a car, the mass, dimensions and dynamic qualities wheel, overhang, base and gauge). To be able to bypass and of which are used when designing a road. Parameters of open the doors of the car, the parking dimensions should the design vehicle, such as dimensions and the minimum be 0.5 m larger than the corresponding dimensions of the turning radius, should be the same for majority of vehicles designing vehicles [12]. of the same class, which are supposed to be used for the The “Methodological guidelines for design and movement of the designed road”. equipment of highways to ensure traffic safety” [13], During the roads and parking spaces designing, it indicated that parking at large recreation areas, at roadside becomes necessary to determine the width of the path of catering establishments, motels and campgrounds should overhang, the size of the manoeuvres space, and geometric be placed between the highway and buildings with vehicle parameters. This in each case requires the construction of separation by types and sizes. Parking areas for trucks the dynamic dimension of the design vehicles, which is a and passenger cars should be demarcated and a separate time-consuming process (especially for the road trains) and entrance to the appropriate temporary parking area should not it is sufficiently mastered by designers. be provided for each type of vehicle. The lack of systematic reference material on these In this case, passenger cars and buses are recommended issues is often the cause of unreasonable design decisions, to be on the left-hand side and trucks on the right in the which either lead to an overestimation of the estimated direction of travel. costs of facilities or do not provide normal operating It is recommended to place the parking of trucks conditions for vehicles. parallel to the axis of movement, while parking of passenger cars mainly should be arranged according to an oblique angle at an angle of 45-60°. For long stays in the 2 Materials and methods parking place, as well as in cramped conditions, when the parking place has one exit, it is recommended to install To determine the width of the manoeuvring of parking vehicles perpendicular to the direction of the axis of spaces, the authors took into account the minimum turning movement. Recommendations are given for the designation radius of the design vehicle and its dynamic clearance. of the average area of coverage for one vehicle, taking into To do this, studies have been conducted that allowed to

VOLUME 22 COMMUNICATIONS 2/2020 34 MIKUSOVA et al.

Figure 4 Manoeuvring schemes for a train (16.5 m2) in a parking: a) the location of the parking space at an angle of 90°; b) the location of the parking space at an angle of 45°; c) the location of the parking space at an angle of 60° (1-way forward; 2- backing; 3- forward ride; 4 - reversing) determine these characteristics [14-16]. In the study, the 3 - Road train drives forward; width of the passage was determined as follows. When 4 - The road train leaves the parking space in reverse. designing the parking space and the entrance of vehicles at It was found that for reversing a large manoeuvring parking spaces, the following schemes and provisions were lane is needed than in other variants. This manoeuvre applied in the calculations: is a common parking method for the road train drivers. 1 - The road train leaves the parking space in the forward With this in mind, the width of the manoeuvring strip was direction; determined. 2 - Auto train drives backward in a parking space;

COMMUNICATIONS 2/2020 VOLUME 22 DESIGNING OF PARKING SPACES TAKING INTO ACCOUNT THE PARAMETERS OF DESIGN VEHICLES... 35

Figure 5 The dependence of the area of parking space on the number of parking spaces for passenger cars: a) and for a train 16.5 m long; b) when set at the corners: 1 - 90°; 2 - 60°; 3 - 45°

The design vehicle made a manoeuvre at the location than at an angle of 60° or 90°. These values do not affect of a parking space at an angle of 90°, 60° and 45° (see Figure the number of parking spaces; moreover, these values are 4). After each manoeuvre, the parking length, manoeuvring the most effective indicator when manoeuvring cars on the lane, and parking width were determined. parking site. Studies have shown that for one passenger car, taking into account manoeuvring, 28.7 m2 of parking space is needed. For a road train length of 16.5 m, this value is 143.1 3 Results and discussion m2 of area. From Figure 5 it follows that with more than five Based on the above data, one can draw the following parking spaces the area of parking space for one vehicle conclusions: does not increase (depending on the angle). When the 1. Less than 5 parking places at an angle of 90° are parking space is located at an angle of 90° and if there economically inefficient for any type of vehicles; are less than 5 parking spaces in the parking, the parking 2. When located at angles of 60° or 45°, up to 4 parking space is reduced by one car. At the location of parking places are considered to be economically inefficient. spaces at angles of 60° and 45°, the indicator is 4 parking The following types of design vehicles were spaces. Similar values are obtained for cars and for trucks. recommended as the most frequently encountered on the Proceeding from this, it can be concluded that, when roads for Russian Federation: passenger car (P); city bus parking places at an angle of 90°, designing less than five (CB); bus (B); articulated bus (AB); truck (T); road train parking spaces is ineffective for any type of car and if consisting of truck tractor and semi-trailer (A16); road placed at angles of 60° or 45°, up to four parking spaces are train consisting of a truck and a trailer (A20). The main considered ineffective. dimensions of the specified design vehicle are given in the From Figure 6 it follows that at the location of parking Table 1 with a template for the design of curves in the plan spaces at an angle of 45° less parking space is required with the minimum radius of the design vehicle: Figure 7

VOLUME 22 COMMUNICATIONS 2/2020 36 MIKUSOVA et al.

b) Figure 6 The required parking area at different angles of installation of vehicles and parking capacity: 1 - one vehicle; 2 - 5 vehicles; 3 - 11 vehicles [a) for passenger cars; b) for road trains with a length of 16.5 m]

Table 1 Recommended size of the design vehicles Dimension, m Designation general overhang Type of design vehicles Wheelbase, m length width front rear RD1) TR2) Passenger car P L 2.90 4.90 1.90 0.90 1.10

City bus CB M2 6.20 12.0 2.50 2.75 3.05

Bus B M3 6.90/1.30 15.0 2.50 2.60 4.20

Articulated bus AB M3 5.96/6.05 18.4 2.55 2.68 3.71

Truck T N3 6.80 12.0 2.50 1.50 3.70

Road train A16 N2+O4 3.80/7.02 16.50 2.50 1.43 2.98

Road train A20 N3+O4 6.80/4.30 19.80 2.50 1.50 0.70 Note: RD1) - vehicle designation adopted in the article. TR2) - designation of cars in accordance with the Technical Regulations “On the safety of wheeled vehicles” (approved by the decision of the Commission of the Customs Union of 9 December 2011 No. 877)

COMMUNICATIONS 2/2020 VOLUME 22 DESIGNING OF PARKING SPACES TAKING INTO ACCOUNT THE PARAMETERS OF DESIGN VEHICLES... 37

Table 2 Minimum turning radius of the design vehicle Type of design vehicles Minimum turning radius, m Minimum outer radius, m Minimum inner radius, m Passenger car (P) 6.55 6.85 4.42 City bus (C) 9.20 10.54 5.40 Bus (B) 10.32 11.52 6.40 Articulated bus (A) 13.12 14.21 10.10 Truck (T) 11.07 11.82 6.15 Road train (A16) 9.69 10.19 6.20 Road train (A20) 12.06 12.63 8.50

Figure 7 Template for the design of curves in the plan with the minimum radius of the design vehicle: a) passenger car (P); b) city bus (CB); c) bus (B); d) truck (T) i) articulated bus (AB); f) road train (A16)

VOLUME 22 COMMUNICATIONS 2/2020 38 MIKUSOVA et al.

Figure 8 Scheme and dimensions of parking spaces for bus: a) when parallel; b) saw tooth; c) perpendicular placement in relation to the driveway or manoeuvring zone. A similar scheme is proposed for other type of buses

Figure 9 Elements of breakdown of parking space for trucks α - β - parking angle; Ø - vehicle installation angle; B - D - manoeuvring strip; C - parking module length

[17], the minimum turning radius - in the Table 2. Scheme design vehicles are presented in Table 4. Considering the and dimensions of parking spaces for bus are given in foreign experience of organizing parking spaces for large- Figure 8. sized vehicles, which provides for entry and exit to parking The resulting sizes of parking spaces and the scheme without reversing (Figure 9), as a result of research, it is of their breakdown are given in Table 3. Recommended recommended to take the dimensions of parking spaces in lengths of parking spaces for longitudinal placement of accordance with values in Table 5.

COMMUNICATIONS 2/2020 VOLUME 22 DESIGNING OF PARKING SPACES TAKING INTO ACCOUNT THE PARAMETERS OF DESIGN VEHICLES... 39

Table 3 Dimensions and average areas of one parking space

Sizes of elements, m Average area for 1 vehicle, m2

Vehicle installation without with angle, degree a b c d e f g h i maneuvering maneuvering

1 2 3 4 5 6 7 8 9 10 11 12 One-way car parking (P) 90 5.0 7.0 17.0 11.5 0.5 6.0 2.5 2.5 0.5 12.5 28.7 60 5.2 4.2 14.6 8.9 0.5 3.2 2.5 2.9 0.5 15.1 25.8 45 4.8 4.0 13.6 8.3 0.5 3.0 2.5 3.5 0.5 16.8 29.0 Two-way car parking (P) 90 5.0 8.0 18.0 12.5 0.5 7.0 2.5 2.5 0.5 12.5 22.5 60 5.2 5.2 15.6 9.9 0.5 4.2 2.5 2.9 0.5 15.1 22.6 45 4.8 5.0 14.6 9.3 0.5 4.0 2.5 3.5 0.5 16.8 25.5 Truck Parking (T) 90 13.0 16.1 42.1 28.6 0.5 15.1 3.5 3.5 0.5 45.5 100.1 60 11.8 12.4 36.0 23.7 0.5 11.4 3.5 4.0 0.5 47.2 94.8 45 10.5 8.7 29.7 18.7 0.5 7.7 3.5 5.0 0.5 52.5 93.5 City bus parking (CB) 90 13.0 16.1 42.1 28.6 0.5 15.1 3.5 3.5 0.5 45.5 100.1 60 11.8 12.4 36.0 23.7 0.5 11.4 3.5 4.0 0.5 47.2 94.8 45 10.5 8.7 29.7 18.7 0.5 7.7 3.5 5.0 0.5 52.5 93.5 Bus parking (B) 90 16.0 19.0 51.0 34.5 0.5 18.0 3.5 3.5 0.5 56.0 120.7 60 14.3 16.1 44.7 29.9 0.5 15.1 3.5 4.0 0.5 57.2 119.6 45 12.4 11.7 36.5 23.6 0.5 10.7 3.5 5.0 0.5 62.0 118.0 Articulated bus parking (AB) 90 19.5 25.1 64.1 44.1 0.5 24.1 3.5 3.5 0.5 68.3 154.3 60 17.3 20.3 54.9 37.1 0.5 19.3 3.5 4.0 0.5 69.2 148.4 45 14.9 18.5 48.3 32.9 0.5 17.5 3.5 5.0 0.5 74.5 164.5 Road train parking (A16) 90 17.5 23.9 58.9 40.9 0.5 22.9 3.5 3.5 0.5 61.3 143.1 60 16.6 18.9 52.1 35.0 0.5 17.9 3.5 4.0 0.5 66.4 140.0 45 13.5 17.4 44.4 30.4 0.5 16.4 3.5 5.0 0.5 67.5 152.0 Road train parking (A20) 90 21.0 33.0 75.0 53.5 0.5 32.0 3.5 3.5 0.5 73.5 187.2 60 18.6 23.8 61.0 41.9 0.5 22.8 3.5 4.0 0.5 74.4 167.6 45 16.0 21.1 53.1 36.6 0.5 20.1 3.5 5.0 0.5 80.0 183.0

Table 4 The length of parking spaces for longitudinal design vehicles

Passenger car Buses Road Train Design vehicle Truck (T) (P) (CB) (B) (AB) (A16) (A20)

Parking space 6.0 14.0 14.0 17.0 22.5 20.0 24.0 length, m

VOLUME 22 COMMUNICATIONS 2/2020 40 MIKUSOVA et al.

Table 5 Sizes of parking spaces at different corners of the parking of trucks

Set angle, degree Sizes of parking spaces, m (see Figure 9)

Ø α β B C D A16 30 30 30 7.5 12.0 7.5 35 35 35 8.5 13.0 8.5 40 40 40 8.7 13.5 8.7 45 45 45 9.5 15.5 9.5 A20 30 30 30 8.0 13.0 8.0 35 35 35 9.0 15.5 9.0 40 40 40 9.2 16.5 9.2 45 45 45 10.0 17.7 10.0

For highways, where trucks and road trains a width of 4.5 meters. The minimum radius for entering the predominate as part of the traffic flow, it is recommended parking lot should be at least 26 m and at the exit - at least to the plan parking spaces with a transverse arrangement so 30 m. Dimensions of the parking spaces at different parking that trucks do not need the turning manoeuvre or movement angles are presented in Table 5. with minimum radii. Design of diagonal parking may vary depending on the angle at which the vehicles are located. To determine the size of the diagonal parking spaces for 4 Conclusions trucks, the authors created a design vehicle model and modelled the movement path of the design vehicle and During the studying of the parking lots sizes the authors determined the dynamic size, as well. With this in mind, found that the previously adopted regulatory documents the main dimensions of the parking lot were determined recommended dimensions for the existing traffic flow in (Figure 9). Using the above approach, the diagonal parking the twentieth century. Regulatory documents contained sizes were determined at an angle from 30° to 45°. The ambiguous, sometimes contradictory information, which is planning of the parking space includes the following contrary to objectives of the road safety and its convenience. actions: alternative placement of parking spaces within Dimensions of the parking space, recommended by the the parking lot, modelling using the dynamic dimensions author, differ from the current regulatory documents in the of the design vehicle of entrances to the parking space and direction of reducing the area for a parking space. This is exits. A necessary element of the traffic organization in the due to the fact that in modern vehicles dynamic indicators parking lot and design of the parking space, is organization are optimized and require less space for manoeuvring. of unimpeded passage of vehicles pass parking spaces with

References

[1] Information and legal portal Garant [online]. [Viewed 2019-27-01]. Available from: http://www.garant.ru/hotlaw/moscow /430367 [2] BURANOV, I. Mayor shifted the burden of responsibility to the region. Kommersant. 2012, 218(5003). ISSN 1563-6380. [3] Monitoring compliance with the norms, rules and standards when designation and construction of roadside facilities (service facilities). Methodical recommendations. Moscow: SIC STSI of the Ministry of Internal Affairs of Russia. 2004. [4] RANKIN, V. U. Automobile transportations and the organization of traffic. Reference book. Moscow: Transport, 1981. [5] BAERWALD, J. E. Transportation and traffic engineering handbook. 3. ed. Washington, D. C.: Institute of Traffic Engineers, 1965. [6] Design and equipment of highways to ensure traffic safety. Methodical recommendations. Moscow: Transport, 1983. [7] ABDUNAZAROV, J. N. Design vehicle for the design of roads (in Russian). Bulletin of the Moscow Automobile and Highway State Technical University (MADI) [online]. 2013, 3(34), p. 81a-84 [accessed 2019-21-11]. ISSN 2079-1364. Available from: https://elibrary.ru/item.asp?id=20164670

COMMUNICATIONS 2/2020 VOLUME 22 DESIGNING OF PARKING SPACES TAKING INTO ACCOUNT THE PARAMETERS OF DESIGN VEHICLES... 41

[8] Regulations for the placement of multifunctional road zones of the service on the highways of the State company “Russian highways” (approved by the order of the State company “Russian highways” June 24, 2013, No. 114) [online]. [Viewed 2018-07-12]. Available from: http://www.rhighways.ru/for_investor/road_service/multifunctionalroad_service_ area [9] POSPELOV, P. I., SHIT, B. A., ABDUNAZAROV, J. N., Designing of parking spaces on car parking (in Russian). Science and Engineers for Roads [online]. 2016, 2(76), p. 6-10 [accessed 2019-21-11]. ISSN 1993-8543. Available from: https://elibrary.ru/item.asp?id=26289975 [10] MIKUSOVA, M., ABDUNAZAROV, J. Modelling of vehicles movements for the design of parking spaces. In: Computational Collective Intelligence. ICCCI 2019. Lecture Notes in Computer Science. NGUYEN N., CHBEIR R., EXPOSITO E., ANIORTE P., TRAWINSKI B. (Eds). Vol 11684. Cham: Springer, 2019. [11] Auto TURN. Advanced vehicle simulations [online]. [Viewed 2018-08-10]. Available from: http://store.softline.ru/ transoft/transoft-autoturn. [12] Help for guests. Guidelines [online]. [Viewed 2019-08-02]. Available from: http://www.gosthelp.ru [13] The Methodical recommendations on designing and equipping the highways for road safety. Minavtodor RSFSR, M.: Transport, 1983. [14] ABDUNAZAROV, J. N. Modelling of the movement of the design vehicles on parking space (in Russian). Science and Engineers for Roads. [online]. 2018, 2(84), p. 10-11 [accessed 2019-21-11]. ISSN 1993-8543. Available from: https://elibrary.ru/item.asp?id=35225761 [15] MIKUSOVA, M., GNAP, J.: Experiences with the implementation of measures and tools for road safety. In: XII Congreso de Ingenieria del Transporte CIT 2016 [online]. Valencia, Spain: Universidad Politechnica de Valencia, 2016. ISBN 9788460899600, p. 1632-1638. Available from: https://doi.org/10.4995/CIT2016.2015.2555 [16] MIKUSOVA, M., ABDUNAZAROV, J., ZUKOWSKA, J.: Modelling of the movement of design vehicles on parking space for designing parking. In: 19th International Conference on Transport Systems Telematics TST 2019: Development of Transport by Telematics [online]. Vol. 1049. Heidelberg: Springer, 2019. ISBN 978-3-030-27546-4, eISSN 978-3-030-27547-1, p. 188-210. Available from: https://doi.org/10.1007/978-3-030-27547-1_15 [17] ABDUNAZAROV, J. N., MAMARASULOVA, M. N., Recommended parameters of design vehicles for the Russian Federation (in Russian). Young Scientist [online]. 2016, 7(2), p. 26-29 [accessed 2019-21-11]. ISSN 2072-0297. Available from: https://moluch.ru/archive/111/27588

VOLUME 22 COMMUNICATIONS 2/2020 42 https://doi.org/10.26552/com.C.2020.2.42-51 ORIGINAL RESEARCH ARTICLE

Irwanda Wisnu Wardhana - Dhani Setyawan - Khairunnisah DETERMINANTS OF LABOR MARKET IN JAKARTA METROPOLITAN AREA: A SURVIVAL ANALYSIS OF COMMUTERS

This study aims to assess the determinants of the labor market in the Greater Jakarta Area (Jabodetabek) with a population of 27.9 million (2010 census) and growth rate of 3.6 percent per annum over the period 2000-2010. With a total area of 4,384 square kilometers (1,693 sq mi), the city has a very high population density of 14,464 people per square kilometer (37,460/ sq mi), while the metro area has a density of 4,383 people/sq km (11,353/sq mi). The paper employs the survival regression analysis by incorporating attributes of commuter, namely gender, age, distance, travel time, wages, stress, education level, double income households and home ownership. The area consists of Jakarta as the receiving labor market and eight municipalities and regencies as labor suppliers. The study utilizes a cross-section data from a commuter survey with more than 4,000 respondents participated using different modes of land transport. The results reveal that some determinants have influenced commuters' resiliency and their willingness to participate in the receiving labor market. This study found that gender, distance, wages and home ownership do not affect the respondent’s decision whether to stay or quit as commuters. On the other hand, the fittest model exhibits that age, education level, stress, travel time and double income households have significant effects on individual's decision to stay or quit as a commuter. It is found that gender, distance, wages and home ownership do not matter for respondent’s decision on whether to stay or to quit as commuters. The model exhibits that age, education level, stress, travel time and double-income household have significant effects on an individual's decision to stay or quit as a commuter. Education level has a positive effect; on the other hand, age, stress, double-income household and travel time have a negative effect. The policy implications for improving the labor supply provision and some contested policy options are suggested, such as the provision of affordable housing in Jakarta, the improvement of commuting enjoyment, the establishment of child care facilities in the office buildings and the creation of more sophisticated jobs within the Jakarta’s surrounding municipalities and regencies. Keywords: commuter, labor supply, survival analysis, transportation

1 Introduction are traveling within, into and out of the city in a day. Of these people, 1.38 million are those who live in the outskirt Jakarta, as Indonesia’s capital city, has been a magnet of Jakarta, such as Depok, Tangerang, Bogor, Bekasi, and to every immigrant to work. Aside from being the center of Banten [2]. every governance activity, Jakarta economic activities also This study would like to assess the determinants of attract more labors, especially from the outside of Jakarta. commuter labor on their decision making to be a commuter Although Jakarta’s economic growth tend to fluctuate, and fulfill Jakarta’s market labor. By employing a survival from 5.97% to 5.93% in the first quarter of 2018, Jakarta has regression analysis, this study will shed light on how contributed 17% to Indonesia’s economy [1]. Higher pay and commuter labors decide to stay or quit as a commuter in more abundant opportunities are some factors influencing the future. individual decision to move to Jakarta. Some of them are willing to travel for 10 or more hours in order to relocate to Jakarta. 2 Literature review Not only attracting those who lived 10 hours away from Jakarta, labor from districts and cities that circled In the last two decades is recorded a very high Jakarta also attracted to gamble in this city, to have a growth of population in developing countries [3]. In line better life through higher pay and a more suitable job that with the growing population, the needs of mobility are match their qualifications. Those who live in the outskirt of also increasing. Accessibility of a sound public transport Jakarta are willing to travel back and forth from their home system has become more important in order to design to their workplace in Jakarta. These people are known as and to evaluate sustainability of a transit system. Besides commuter labors. There are 2.43 million commuters who the transport system, access for public transport has been

Irwanda Wisnu Wardhana1, Dhani Setyawan1,*, Khairunnisah2 1Fiscal Policy Agency, Indonesia Ministry of Finance, Jakarta, Indonesia 2Statistics Indonesia, Jakarta, Indonesia *E-mail of corresponding author: [email protected]

considered to have an impact to life satisfaction [4-5]. If the In a recent study, [14] were analyzed the factors high demands of public transports were not able to meet affecting women’s decisions in choosing the transport a certain level of supply of transport infrastructure that mode in order to determine individual travel requirements may lead to congestions and increasing of waiting times that are affected by socio economic condition, life stages in the streets and public transport hubs [6]. Improvements and gender. Their study employed 6000 household’s data in the public transport systems lead to betterment in other and investigated mobility behavior data provided by Zilina services sectors [7]. Thus, developing and formulating an self-governing region from 2013 to 2016. Authors found that efficient public transport system have become one ofthe the relation of daily travel mode, characteristics of activity primary objectives for planners and policy makers in many participation, and socio-demographics of male and female metropolitan cities across countries [8]. commuters were more gender sensitive. Many studies have investigated several correlation and impact of accessibility of public transport to the life satisfaction and environment that could cause a certain 2.1 Influence of commuters on labor market supply impact on quality of life and public health [4]. Improvement in the sustainability of public transport, public health, An individual decision on doing the commuting can economic conditions of residents and environments can be caused by wages creating the process in the labor be achieved by reducing the use of private transport and market that differs between areas. The amount of wages shifting to the mass transport system and environmentally received by commuters is assumed to be the basis for friendly transport such as cycling and walking [9]. them to choose transportation modes [15]. In this study, The relation between job opportunities and Jakarta’s commuters that fill the labor market in 2014 have accessibility for public transport has become an interesting been attracted to the minimum wage policy that fixated topic to be investigated in many studies. One of the on IDR 2,44 million. It was the highest minimum wage purposes of enhancing urban public transport systems to be set compared to the other eight cities/regencies is to improve commuting and also to facilitate individual outside Jakarta. Other than Jakarta’s minimum wages, performance on the employment market [4]. According to transportation cost and housing price have accounted on [10], a positive correlation between labor force attachment household’s decision making on whether they should be a and transportation accessibility might not be caused by the commuter or not. Social demographic characteristics, such public transport role, but is a result of spatial urban factors. as education level, age, gender and race also influenced an Enhancing job accessibility might increase earnings and individual decision on commuting [16]. individual employment rates by various mechanisms [4]. Labor supply limitation due to migration will keep Furthermore, according to [11], the correlation between minimum wage high in the receiving labor market [17]. employment and transport accessibility has mostly become Meanwhile, commuters that are eager to increase their an area of dispute in formulating transport infrastructure utilization to get the most satisfying pay and the job will and geographic conditions. It seems that the job locations travel back and forth in one day. This event will imply the and the workforce’s residential locations could be overcome unlimited labor availability in the city, additionally with a by providing a sound transit system. Labors might not regional minimum wage policy set above other areas than consider various relevant job opportunities due to long Jakarta. Commuters flow will improve labor supply for the commuting time [12]. In this sense, the characteristics of job in the commuters’ receiving labor market. Hence, it will a sound public transportation system can become one of lower the cost of opening new job opportunities. Okun’s productive strategy to challenge unemployment [13]. Law has stated that decrease in labor demand on receiving The relation between employment and access of labor market will affect in the rise of product and services transit system has been revealed by [13] which employed demand, assuming that commuters utilize and spend their geographical information system to investigate the various money in the district where they work, not where they labors characteristics within different transport systems. originated from [16]. This study found that the transport accessibility has become one of the key factors in determining employment participation. Further, by evaluating the correlation 2.2 Staying duration on current living area between labor participation levels and transport access and employing the two-stage least squares regression model, Sociodemographic characteristics affect an individual [13] proved that a stronger relationship between labor decision on doing the commuting. How long they have participation and public transit accessibilities. Further, a stayed in their current living area is one of the satisfaction study of [11] justified the previous study of [13]. In [11] indicators on migration process they perform, because it was investigated the available network of public transport can describe the adaptation process of the immigrant to system influence on the employment rate variation at a local their environment [18]. Recent migration has limited staying level and a statistically significant relationship between the duration up to five years compared to their current living shorter commuting time and higher employment levels was area. For this reason, this study was aimed to find how the found. personal satisfaction to keep or stop being a commuter, regarding their duration of stay in their current living

VOLUME 22 COMMUNICATIONS 2/2020 44 WARDHANA et al.

area. Staying duration will describe individual resilience Survival analysis, in particular, the survivor function on shortening their traveling distance to the workplace. and hazard ratio will measure the commuters’ resiliency. The longer the staying duration, it will describe different This research uses parametric survival regression with log individual decision on comfortability factors that cannot be logistic distribution. Survivor function S(t) is a function offered by other areas. that stated individual chances to stay longer in their current living area and become a commuter to Jakarta up until the certain point of time, which can be defined as: 2.3 Commuting duration, distance and double- income household migration strategy St = PT$ t . (1) ^ h ^ h Every individual has own resilience stability to perform The random variable T defined staying duration in commuting in a day for one hour [19]. If the commuting the current living area and keep being a commuter. The takes place more than an hour, it will affect individual probability density function f(t) is: decision making to stay as a commuter or to stop the act dS t of commuting. Male commuter and female commuter ft =- ^ h , (2) ^ h dt have different resilience in commuting; where male are described as more resilient on long-distance commuting, proof: this is not what happens with female commuters. Female 3 commuters’ resilience can be affected by the household st = # fxdx , (3) ^^hht conditions. Household where husband and wife are both t 3 working outside their original labor market will discuss ##fxdx +=fxdx 1 , (4) -3 ^^hht whether they will stay as commuters or quit the job. A t tendency to quit a job more likely happens to wives due to # fxdx =-1 St , (5) -3 ^^hh household conditions, the school-age child they have would make a female more likely to stop commuting and choose a t df# x dx dS1 - t job much closer to her home [20-21]. : -3 ^ h D = , (6) dt 6 dt ^ h@ Additionally, the experience of having a long-distance dS t commuting with a longer duration will affect an individual ft =- ^ h . (7) ^ h dt decision on staying or quitting as a commuter [22]. A study by [19] on measuring commuting resilience duration with On the other hand, the Hazard function is used to survival analysis, while [22] divide subsamples of the define the probability of someone to keep staying in their research based on how long they have been commuting current living area and stop being a commuter in time t with multinomial analysis. In this study, the research team subject to the provision of someone quitting as a commuter has used the survival analysis on the duration of staying in up until the time they are willing to quit. This study used 5.9 their current living area with individual decision to stay or years as the limitation based on survival time mean. Given stop being a commuter. is the function:

Pt#$Tt1 + dtT t ht = lim ^ h . (8) ^ h dt " 0 dt 3 Methodology

This research uses Jabodetabek commuters in 3.1 Relation between Survivor function and Hazard 2014 survey data that has been conducted by Statistics function Indonesia. This survey has been conducted in five districts in Jakarta, namely South Jakarta, East Jakarta, Central Pt# Tt1 + dt Jakarta, West Jakarta, North Jakarta and eight surrounding Pt#$Tt1 +=dtT t ^ h (9) ^ h PT$ t cities and regencies, namely Bogor Regency, Bogor City, ^ h Bekasi Regency, Bekasi City, Depok, Tangerang Regency, Ft+-dtFt Tangerang City and South Tangerang City. This research = ^ h ^ h, (10) St used 1,712 households whose household members perform ^ h Pt#$Tt1 + dtT t 1 commuting to Jakarta in one day back and forth to fulfill ht = lim ^ h $ , (11) ^ h dt " 0 ' dt 1 St Jakarta’s labor market. The total commuter individuals, who ^ h ft headed to Jakarta, were 3,522. To analyze the commuting ht = ^ h (12) ^ h St resilience of the individual, 2,485 units sample, who has ^ h dS t been living in their current living area up to 10 years, were because t =- ^ h , (13) ^ h dt gathered. Ten years of living decided as the limitation d because many cases of quitting as a commuter happened therefore ht =- log St , (14) ^^hhdt " , in this period, although it is still an indifferent duration for commuters to stay or quit.

COMMUNICATIONS 2/2020 VOLUME 22 DETERMINANTS OF LABOR MARKET IN JAKARTA METROPOLITAN AREA: A SURVIVAL ANALYSIS... 45

log log St ==htdt Ht , (15) 3.5 The log-linear format from Accelerated Failure ^^hh# ^ h Time log-logistic model logeSt xp -Ht . (16) ^^hh^ h Log linear is a group of a variable in surviving time Ti: The parametric method, applied in this study, is using analysis method with log logistic distribution assumption, log Txii=+na1ip++g avx pi + fi . (26) after distribution test conducted with Matlab software. Thus, Survivor function and log-logistic distribution used in Using the Survivor function definition, thus: this study are defined as:

St ==PT$$tPloglTti og = eti k ^ h ^^h h St =-11Ft =- i k = log tx--nvg - tti (27) ^^hh1 + et P fi $ . (17) b v l 1 +-etiikket 1 = = . 1 + eti k 1 + eti k Assuming that f distributes logistics, the density of Meanwhile, Hazard function with log-logistic opportunity function probability and Survivor function distribution will be: from f are:

ft ketiikk-1 1 + et ef ht ==^ h = . (18) f f = , (28) ^ h St 1 + eti k 2 1 ^ h 1 + ef 2 ^ h ^ h ^ h 1 The Accelerated Failure Time (AFT) log-logistic model S f = . (29) ^h 1 + ef was used to analyze the survival probability.

Therefore, the Survivor function from Ti is: -1 log tx--nvg - tti 3.2 Accelerated Failure Time (AFT) log-logistic Sti =+1 exp . (30) ^h < ( 2F model The Survivor function, with i-individual distributes the

This model describes when the commuter stops log-logistic with ih+ kki, parameter estimation, where ^h happening faster than the survival time (accelerated failure hi is: time).

hbip=+11xxiibb22++g x pi . (31)

3.3 Hazard function with the AFT model The Survivor function given from T is:

hhii- 1 hti = eh0 et, (19) Sti = . (32) ^ h ^ h ^h 1 + etih+kki ht0 = Initial Hazard Function, (20) Compared to T Survivor function that: ^h n hti =-InitialHazardFunctiononi, (21) i =- , (33) ^h v 1 hbii=+11xxbb22ii++g ppx , (22) k = v . (34)

where i=1,2,…,n. The Hazard ratio from the Proportional Odds Log- logistic model is:

hti hi -1 3.4 Survivor function with AFT model ^ h =+11eS- 0 t . (35) ht0 6 ^ h ^ h@ ^ h fi t Sti = ^ h , (23) ^ h hi t ^ h

SSi tI=-nitialuurvivorf nction on i, (24) 4 Results and discussion ^h hti =-Hazarduf nctiononi. (25) When forming its labor market, Jakarta as Indonesia’s ^h capital city cannot separate the local labor (stayers’ labor) How long commuter will stay in his current living area and labor who are mobile (movers’ labor) with daily and keep being a commuter is estimated by ih+ kki, duration (commuting labor) or more than one-day duration ^h parameter. (circular labor) in six months period. The increasing percentage of Jakarta, Banten and West Java’s commuter labors indicated a spatial interaction to fulfill the labor market for each area. The Spatial interaction happened

VOLUME 22 COMMUNICATIONS 2/2020 46 WARDHANA et al.

Table 1 Data descriptions Log Logistic Survival Variable 1 = Quit 0 = Stay Total Sample n=2485 Gender Male 1 Female 2 Age Education Up to Elementary/Islamic Elementary or equivalent 1 Junior High School/Islamic Junior High School or equivalent 2 Senior High School/Islamic Senior High School or equivalent 2 Diploma I/II/III 4 Diploma IV/Bachelor/Master/Doctoral 5 Homeownership status Self-owned 1 Rent/Contract 2 Other 3 Feeling stress when headed to/from receiving labor market Yes 1 No 2 Commuting Duration Mileage Average income from the primary job Double-income household Yes 1 No 2 in the form of labor mobilization that moves across their place with spatial distribution pattern that spread more city or district border. This interaction implies the increase widely, along with the growth of metropolitan cities. Table of income per capita on receiving labor market and an 1 below describes the characteristics of respondents in this increase in welfare on the labor’s origin area, especially study. for a household with commuter labor. This phenomenon Labor mobility that happens with a daily duration can describe two sides of commuter labor as an input on from the surrounding of Jakarta is an implication caused production factor in the receiving labor market to produce by suburbanization around Jakarta. 12.44% of labors from output and as a resident of the area of origin with better Depok fulfilled Jakarta’s labor market and became the most income level compared to when they do not move. In [15] significant percentage compared to the other 13 districts/ is explained that the commuter labor is most likely to work cities/regencies that directly bordered with Jakarta. 32.45% outside their current living area than in their origin area, of the commuter labor from 13 districts/cities/regencies assuming reservation wage for every labor is treated as around Jakarta, headed to South Jakarta and Central equal. Jakarta every day. That describes how Jakarta’s biggest This study has analyzed how the commuter labor’s economy happened in the Central and South Jakarta, characteristics from the surrounding area of Jakarta, such supported by commuter labors from 13 districts/cities/ as Banten district and West Java, are fulfilling Jakarta’s regencies. The Central and South Jakarta are the biggest labor market, based on Jabodetabek commuter survey in Jakarta’s most significant GDP contributors, with 24.34% 2014 by Statistics Indonesia. Jakarta, as a megapolitan, has contribution come from Central Jakarta and 22.35% from grown through the suburbanization process that happens South Jakarta. Moreover, commuter labors that headed to in the surrounding area of Jakarta. This phenomenon, Jakarta fulfill governance and public service sector (21.78%), described by [23], happened where suburbanization takes trade sector (19.85%), insurance and finance (12.21%) and

COMMUNICATIONS 2/2020 VOLUME 22 DETERMINANTS OF LABOR MARKET IN JAKARTA METROPOLITAN AREA: A SURVIVAL ANALYSIS... 47

Table 2 Data analysis Total Samples Full Model Fitted Model (n = 2485) Variable Willing to Quit Coefficient Coefficient Yes No Samples 173 2312 Gender Male 126 1619 0.0185968 Female 47 693 (0.1262742) 0.0139384 ** 0.0161032 ** Age 173 2312 (0.0065005) (0.0063751) Education Level Elementary/Islamic Elementary or equivalent 14 100 Junior High School/Islamic Junior High School or 0.6434834 ** 0.6277425 ** 11 169 equivalent (0.2849675) (0.2879138) Senior High School/Islamic Senior High School or 0.5567855 *** 0.5765493 *** 94 1122 equivalent (0.2094445) (0.2113748) 0.8876308 *** 0.9076511 *** Diploma I/II/III 11 214 (0.2759957) (0.2778538) 0.7377673 ** 0.7590284 *** Diploma IV/Bachelor/Master/Doctoral 43 707 (0.2291076) (0.2267885) Home ownership status Self-owned 123 1723 -0.1110324 Rent/Contract 32 380 (0.1402029) -0.1814356 Other 18 208 (0.1768024) Feeling stress when heading toward/ from the activity area

111 1058 Yes

0.3410174 *** 0.3290216 *** No 62 1254 (0.1089175) (0.1091749) -0.003817 ** -0.0023322 * Commuting Duration 173 2312 (0.0016782) (0.0013896) 0.0067719 Mileage (Distance) 173 2312 (0.004922) The average income per month from the primary job -1.58e-08 (7.72e-09) Double-Income Household -0.1936276 * -0.1963347 * 65 833 Yes (0.1121168) (0.1101553

No 108 1479 1.953655*** 1.886609 *** Constants (0.3370987) (0.3215616) -0.8366013 *** -0.826973 *** 1/ln_gamma (0.0916982) (0.0918275) 0.4331803 0.4373712 gamma (0.0397219) (0.0401627) * p<0.10, ** p<0.05, *** p<0.01

VOLUME 22 COMMUNICATIONS 2/2020 48 WARDHANA et al.

Table 3 Surviving probability and Hazard ratio on quitting as a commuter Survivor Function Hazard Function Hazard Variable Stay as a Commuter Quit as a Commuter Ratio 0 i 0 i i Global 0.466 -0.109 Age 0.471 -0.108 0.992 Education Level Elementary/Islamic Elementary or equivalent Junior High School/Islamic Junior High School or equivalent 0.799 -0.041 0.376 Senior High School/Islamic Senior High School or 0.876 -0.025 0.232 equivalent Diploma I/II/III 0.986 -0.003 0.026 Diploma IV/Bachelor/Master/Doctoral 0.989 -0.002 0.022 Feeling stress when heading toward/from the activity area Yes No 0.659 -0.070 0.639 Commuting Duration 0.466 -0.110 1.002 Double-Income Household Yes 0.488 -0.105 0.960 No manufacturing industry (11.73%). These four sectors are live in their current place following where their husband/ the most significant contributor to Jakarta’s gross domestic, wife/parents/kids live. 11.19% of them said it is because of local product and become the main characteristic of a big housing and followed by working reason 8.21%. Therefore, city such as Jakarta. In 2014, the trading sector contributed it contributed to individual decision to stay in their current 16.64% to Jakarta’s GDP, followed by governance and home or choose to seek a home near their workplace and public service sector (15.18%), manufacturing (12.94%) and decide to quit as a commuter. financial services (10.21%). Commuting duration and previous experience as The technology development, marked by better an immigrant are the reason why individual decided to transportation system, which connects Jakarta and its maximize their utilization to stay in their current area. surroundings, has affected the commuter labors to mobilize 30.72% commuters explained that they decided to commute and head toward to Jakarta without any significant time because of the security and convenience in their current constraint. According to [19], a stable individual will travel living area, while 13.49% commuters are willing to sacrifice to his receiving labor market in 60 minutes duration. their two hours to go to work for their current living area. Additionally, 40.60% commuter labor who headed to On the other hand, Jakarta’s labor market has become the Jakarta, are more likely to travel in 31-60 minutes duration reason for 34.64% commuters to get the job that matches and 16.7% travel for 11-20 km from their home to their their skills and competence. Table 2 below describes the workplace. Travel duration can be an indicator to describe results of the data analysis. the commuter labor supply from the surrounding area that 6.39% stated that they want to make a better living fills the Jakarta’s labor market. through higher income, compared to their current living Jakarta, where most of the economic activities area. That can be an indicator what happens in the labor happened, has attracted labors from outside of Jakarta to mobility where the opportunity of getting higher income take a role in the labor market. Most of Jakarta’s commuters in receiving labor market is more significant than in their come from the surrounding area of Jakarta and become a current living area [24]. recent immigrant, not a resident. Experience on becoming Rational individual choice is more likely to lean to an immigrant, where they have long experience to live in the decision to avoid long commuting time [22, 25]. Thus, their current area that differs from where they lived five every individual will maximize their utilization when facing years ago, is an indicator of immigration satisfaction [18]. two choices between stay or quit as a commuter. This 8.92% of commuters have lived in their current area for five research used willingness to quit as a commuter for the years or different home compared to five years ago. While next one-year approach to limit the event on analyzing 57.04% have experienced living in the area other than their commuter resilience on fulfilling Jakarta’s labor market. current districts/cities/regencies, 19.39% of the commuters According to Jabodetabek commuter survey in 2014, 70.56%

COMMUNICATIONS 2/2020 VOLUME 22 DETERMINANTS OF LABOR MARKET IN JAKARTA METROPOLITAN AREA: A SURVIVAL ANALYSIS... 49

commuter decided to stay in their current living area for shown a declining hazard value. Thus, it can describe how less than or up to 10 years. Therefore, this study takes that individual choose to stay as a commuter. sub-population to see their commuting resilience to their In the end, Jakarta’s government can apply a policy decision on quitting as commuter labor. Table 3 below that will comfort commuters and help them to stay as a indicates results of the commuter decision. commuter. Nearly one-third of the commuter labors fulfilled In Table 3 is shown that commuter probability of staying Jakarta’s labor market and contributed to Jakarta’s a commuter is more significant than the hazard of quitting economy. Commuter labors resilience can be drawn by as a commuter. Based on the education level, Jakarta has examining their commuting duration. According to [19], 60 offered much more job opportunity for those who earn minutes that every labor takes every day to commute is a education higher than senior high school level compared description of labor’s duration stability. It describes how the to the local area. Therefore, the higher the education level, individual resilience on commuting in one day. Therefore, if the more commuters will stay as a commuter rather than commuting is done repeatedly more than one hour a day, it quit. In order to keep their labor working in their origin will deflate the individual’s utilization. Thus, according to area, commuter’s area of origin must develop economic [25], the individual will tend to find a home closer to their activities that will expand jobs opportunity for labor with workplace. higher education level qualification. Additionally, a regional Staying duration in the current living area can describe minimum wage that Jakarta has established already matches the adaptation process every recent migrant is going the commuter’s utility. It can be seen through commuter’s through, where they had a different living area with what probability to stay rather than to quit as a commuter. they have today. Staying duration is also closely related Based on age, the hazard ratio was found to be higher to immigrant’s psychological wellness [26]. When the than the opportunity to stay as a commuter. The older is adaptation process is not going well and the burden the commuter, the more likely they want to work in their of commuting duration is more significant than before, origin area and fulfill their local area labor market, rather commuters may be affected by a psychological disorder, than being a commuter in the future. This is also the case which in this study can be described with stress and plan of a double-income household where the probability of to quit on being a commuter in the next year. Meanwhile, one of the household members is more likely to quit as a a household with double income where husband or wife commuter is more significant due to the internal decision in work, and both are commuters will be more likely to face the household, regarding the family condition and where the a tendency to decide which one will quit as a commuter. In school-aged kid is in the household or not. Thus, the female majority of cases, women tend to quit as a commuter [22]. commuter is more likely to fill the local’s labor market A study by [22] described that the duration of being rather than being a commuter [22]. For the commuting a commuter could be affected by education level and duration, the longer the commuting takes time, the higher gender. While age does not describe someone’s willingness the chance of someone to quit as a commuter. However, this to quit as a commuter or have a shorter duration on is not the case where commuter does not feel any stress on commuting, it is still an aspect that needs to be considered commuting. The more commuter resistance to stress, the regarding the ability to perform commuting. Related to a probability of quitting or staying is almost the same. commuting distance where housing location is formed in Jakarta offers significant job opportunities with tight the surrounding area of Jakarta, home ownership status competition. It has attracted commuter labor to participate becomes an essential consideration for someone on in Jakarta’s labor market. Productive age with higher level deciding to have a home closer to their workplace and quit education commuter is more likely to contribute to Jakarta’s being a commuter. labor market. On the other hand, to maintain the large This research used survival analysis in seeing numbers of commuters to Jakarta’s economy, the public commuter resilience on commuting with current living transportation system must be able to make commuters feel area approach as a stimulus to migrate, shortening their at ease on their commuting time so that long commuting commuting duration from home to workplace based on time will not be a constraint for a commuter to keep on their commuting adaptability to their environment. commuting. Based on Table 2, it describes that gender, distance, income and home ownership do not affect the respondent’s decision whether to stay or quit as commuters. The fittest 5 Conclusions model exhibits that age, education level, the psychological condition, commuting duration and availability of spouse With the growing economy activities in Jakarta, the who is also a commuter can affect individual’s decision to surrounding area of Jakarta will eventually get spillover stay or quit as a commuter. from this event. One of the aspects that is most likely to This research used parametric survival regression be affected is the labor market. With the development where gamma value was defined as a shape from data towards a better transportation system, housing price and distribution that describes the hazard value. The Survival opportunities available in the megapolitan city, it surely will model on the duration of staying in Jakarta’s commuter has attract more labors from outside of Jakarta to participate in Jakarta’s labor market. However, several factors are playing

VOLUME 22 COMMUNICATIONS 2/2020 50 WARDHANA et al.

a role in commuter labor decision making, whether they most people are still reluctant to work in the surrounding want to stay or quit as a commuter. area of Jakarta due to the small number of job opportunities The policy implications for improving the labor supply given by the local labor market. If Jakarta’s surrounding provision and some contested policy options are suggested area wants to retain their residents to work in their area, such as the provision of affordable housing in Jakarta, the jobs opening, especially for labor with a higher level of creation of jobs within the Jakarta’s surrounding cities and education, must be attractive enough to the employee improvement of the commuting enjoyment. candidates. It can be started by giving better incentives and Labors with a higher educational level are those who more job opportunities that match their qualifications and are willing to stay as a commuter in order to pursue their interests. most wanted and most suitable job for their qualification. If Jakarta wants to retain their numerous commuter Jakarta has offered much more opportunities compared to labors, it must improve the transportation system. Not its surrounding areas such as Depok, Bogor, Tangerang and only the inner-city transportation that must be developed Banten. However, as labor grow older, they are more likely but the intercity transportation, as well. Therefore, it will to stay in their area of origin and plan to quit as a commuter, make commuters more likely to head to Jakarta because especially if the commuting time is longer than one hour there will be no more significant constraint to head to and trigger stress on the labor. This is also the case of work, especially the time constraint. Tariff reduction on women commuter, especially if they already have a spouse public transportation can also play a role to retain and and a school-aged kid, where they are more likely to quit as attract more labor from the surrounding area. In addition, a commuter due to household decision making. the enjoyment during the traveling time must be taken Jakarta surroundings already has its advantage of into consideration. A seamless experience for commuters, the lower house pricing compared to Jakarta. Thus, the when changing their transportation modes would be very provision of affordable housing in Jakarta will help retaining much welcomed. Additionally, entertainment during the labor supply. However, the lower housing prices and rents commuting is also useful to reduce boredom such as access would be a trigger to labors to be commuters. Furthermore, to free internet.

References

[1] TEMPO, 2019. BI: Jakarta Contributes to 17 Percent of Indonesia`s Economy - Tempo [online]. [Viewed 2019-11-05]. 2018. Available at: https://en.tempo.co/read/922840/bi-jakarta-contributes-to-17-percent-of-indonesias-economy [2] BUDIARI, I. 1.38 million commute into Jakarta daily. The Jakarta Post [online]. 2015 [accessed 2019-11-05]. ISSN 0215-3432. Available from: https://www.thejakartapost.com/news/2015/02/17/138-million-commute-jakarta- daily.html [3] BUHAUG, H., URDAL, H. An urbanization bomb? Population growth and social disorder in cities. Global Environmental Change[online] . 2013, 23(1), p. 1-10. ISSN 0959-3780. Available from: https://doi.org/10.1016/j.gloenvcha.2012.10.016 [4] SAIF, A., ZEFREH, M., TOROK, A. Public transport accessibility: a literature review. Periodica Polytechnica Transportation Engineering [online]. 2018, 47(1), p. 36-43. ISSN 0303-7800. eISSN 1587-3811. Available from: https://doi.org/10.3311/PPtr.12072 [5] KONECNY, V., BEREZNY, R., PETRO, F., TRNOVCOVA, M. Research on demand for bus transport and transport habits of high school students in Zilina region. LOGI - Scientific Journal on Transport and Logistics [online]. 2017, 8(2), p. 47-58. ISSN 2336-3037. Available from: https://doi.org/10.1515/logi-2017-0016 [6] SAMEK LODOVICI, M., TORCHIO, N. Social inclusion in EU public transport. Directorate-general for internal policies [online]. European Union, 2015. Available from: http://www.europarl.europa.eu/studies [7] ABREHA, D. A. (2007). Analysing Public Transport Performance Using Efficiency Measures and Spatial Analysis: The Case of Addis Ababa, Ethiopia. Thesis Report, International Institute for Geo-information Science and Earth Observation, Enschede, the Netherlands. [8] SAGHAPOUR, T., MORIDPOUR, S., THOMPSON, R. G. (2016). Public transport accessibility in metropolitan areas: A new approach incorporating population density. Journal of Transport Geography. 54, pp. 273–285. https://doi.org/10.1016/j.jtrangeo.2016.06.019 [9] ELIAS, W., SHIFTAN, Y. The influence of individual’s risk perception and attitudes on travel behavior. Transportation Research Part A: Policy and Practice [online]. 2012, 46(8), p. 1241-1251. ISSN 0965-8564, eISSN 1879-2375. Available from: https://doi.org/10.1016/j.tra.2012.05.013 [10] KORSU, E., WENGLENSKI, S. Job accessibility, residential segregation and risk of long-term unemployment in the Paris region. Urban Studies [online]. 2010, 47(11), p. 2279-2324. ISSN 0042-0980, eISSN 1360-063X. Available from: https://doi.org/10.1177/0042098009357962 [11] JOHNSON, D., ERCOLANI, M., MACKIE, P. Econometric analysis of the link between public transport accessibility and employment. Transport Policy [online]. 2017, 60, p. 1-9. ISSN 0967-070X, eISSN 1879-310X. Available from: https://doi.org/10.1016/j.tranpol.2017.08.001

COMMUNICATIONS 2/2020 VOLUME 22 DETERMINANTS OF LABOR MARKET IN JAKARTA METROPOLITAN AREA: A SURVIVAL ANALYSIS... 51

[12] PHILLIPS, D. C. Long commutes or neighborhood perceptions: Why do employers avoid applicants from high- poverty neighborhoods? [online]. Available from: http://poverty.ucdavis.edu/sites/main/files/file-attachments/ phillips_proposal_web.pdf [13] SANCHEZ, T. W. The connection between public transit and employment. Journal of the American Planning Association [online]. 1999, 65(3), p. 284-296. ISSN 0194-4363, eISSN 1939-0130. Available from: https://doi.org/10.1080/01944369908976058 [14] POURHASHEM, G., BUZNA, L., KOVACIKOVA, T., HUDAK, M. Exploring women travel behaviour in the region of Zilina from large scale mobility survey. In: Reliability and Statistics in Transportation and Communication RelStat 2018 : proceedings. KABASHKIN I., YATSKIV (JACKIVA) I., PRENTKOVSKIS O. (eds) [online]. Lecture Notes in Networks and Systems. Vol 68. Cham: Springer. ISBN 978-3-030-12449-6, eISBN 978-3-030-12450-2. Available from: https://doi.org/10.1007/978-3-030-12450-2_10 [15] ROUWENDAL, J. Search theory and commuting behavior. Growth and Change [online]. 2004, 35(3), p. 391-418. eISSN 1468-2257. Available from: https://doi.org/10.1111/j.1468-2257.2004.00254.x [16] RUSSO, G., TEDESCHI, F., REGGIANI, A., NIJKAMP, P. Commuter effects on local labor markets: A German modelling study. Urban Studies [online]. 2014, 51(3), p. 493-508. ISSN 0042-0980, eISSN 1360-063X. Available from: https://doi.org/10.1177/0042098013498281 [17] SAMUELSON, P. A. Economics. New York: McGraw-Hill, 1964. [18] NETO, F., FONSECA, A. M. The satisfaction with migration life scale. International Journal of Intercultural Relations [online]. 2016, 54, p. 47-54. ISSN 0147-1767. Available from: https://doi.org/10.1016/j.ijintrel.2016.07.004 [19] JOLY, I. Stability or regularity of the daily travel time in Lyon? Application of a duration model. International Journal of Transport Economics. 2006, 33(3), p. 369-400. ISSN 0303-5247, eISSN 1724-2185. [20] GREEN, A. E. A question of compromise? Case study evidence on the location and mobility strategies of dual career households. Regional Studies [online]. 1996, 31(7), p. 641- 657. ISSN 0034-3404, eISSN 1360-0591. Available from: https://doi.org/10.1080/00343409750130731 [21] SANDOW, E. Commuting behavior in sparsely populated areas: evidence from northern Sweden. Journal of Transport Geography [online]. 2010, 16(1), p. 14-27. ISSN 0966-6923, eISSN 1873-1236. Available from: https://doi.org/10.1016/j.jtrangeo.2007.04.004 [22] SANDOW, E., WESTIN, K. The persevering commuter - duration of long-distance commuting. Transportation Research Part A: Policy and Practice [online]. 2010, 44(6), p. 433-445. ISSN 0965-8564, eISSN 1879-2375. Available from: https://doi.org/10.1016/j.tra.2010.03.017 [23] HENDERSON, V. Medium size cities. Regional Science and Urban Economics [online]. 1997, 27(6), p. 583-612. ISSN 0166-0462. Available from: https://doi.org/10.1016/S0166-0462(96)02169-2 [24] BORJAS, G. J. Labor economics. New York: McGraw-Hill Education, 2013. ISBN 978-0073523200. [25] ZELINSKY, W. The hypothesis of the mobility transition. Geographical Review [online]. 1971, 61(2), p. 219-249. ISSN 0016-7428, eISSN 1931-0846. Available from: https://doi.org/10.2307/213996 [26] WARD, C., BOCHNER, S., FURNHAM, A. The Psychology of culture shock. 1. ed. East Sussex: Routledge, 2001. ISBN 0-415-16234-3.

VOLUME 22 COMMUNICATIONS 2/2020 52 https://doi.org/10.26552/com.C.2020.2.52-59 ORIGINAL RESEARCH ARTICLE

Damian Wierzbicki - Kamil Krasuski METHODS OF PREDICTING THE HEADING, PITCH AND ROLL ANGLES FOR AN UNMANNED AERIAL VEHICLE

The article discusses handicaps in predicting values of rotation angles with regard to Heading, Pitch and Roll for an Unmanned Aerial Vehicle. Within the simulation of the rotation angle values, the linear, polynomial and logarithmic methods were used. The programme source code was written in the numerical editor Scilab 5.4.1. The source data for investigation were recorded by a measuring device Trimble UX-5. The article provides results of comparing the real values of Heading, Pitch and Roll rotation angles to findings obtained from the prediction methods. Based on the conducted research, it was found that the largest value of standard deviation parameter in prediction of the rotation angles is for the angle of Heading, as it equals approximately 5o, whereas the smallest ones are for the Roll and Pitch angles, equalling less than 1.4o. Keywords: UAV, prediction, heading, pitch, roll

1 Introduction flight stage of an aircraft, exploiting available numerical tools and mathematical algorithms. A fundamental navigation aspect in the Unmanned The primary objective of this article is a possibility for Aerial Vehicle (UAV) technology is determination of presenting the methods of prediction of the HPR angles, coordinates and orientation angles in airspace. In the case as well as a verification of their results in comparison to of determining the UAV coordinates, the preferred method the actual data, registered by an UAV. For this purpose, of positioning is the GNSS satellite technology [1], using real readings of the HPR angles from the Trimble 5 code observations on the L1 frequency [2]. However, the platform were taken. In the simulation calculations, three UAV orientation during a flight mostly relies on readings methods of predicting the HPR values were used, i.e. the from the IMU device, being part of the INS sensor. The method of linear regression, the polynomial method and the IMU device has in-built accelerometers (to determine the logarithmic method. aircraft acceleration along the three axes of the reference system) and gyroscopes (to determine three angles of rotation). Ultimately, the INS system provides six degrees 2 Methods of research of freedom within the navigational designation of the user’s position with an integration based on GPS/INS data [3]. In the framework of simulation of predicting the The Heading, Pitch, Roll (HPR) rotation angles refer to the rotation angles, i.e. Heading, Pitch and Roll, three test internal aircraft system, representing the UAV turn, bank methods were exploited, i.e.: and rotation [4]. An important element in determining the • the linear regression method, HPR angles of rotation is also a simulation of predicting • the polynomial method, the UAV in-flight space orientation. This issue is crucial as • the logarithmic method. it allows specifying the HPR angular values for an UAV in The model of linear regression is described by the following cases: dependence [5]: • an entire loss of data from an UAV, • loss of UAV control due to mechanical damage or Ya=+$ Xb, (1) forces of Nature, • failure of a data link from an UAV to the flight operator, where: • loss of visual contact with an UAV. Y - stands for the value of the Heading, Pitch or Roll rotation The fundamental task of the user, when taking an angle, advantage of the mechanisms of an UAV orientation a - is a determined linear coefficient, prediction, is to restore, as closely as possible, the missing X - indicates the number of the next measurement epoch, b - is a determined linear coefficient.

Damian Wierzbicki1,*, Kamil Krasuski2 1Institute of Geospatial Engineering and Geodesy, Faculty of Civil Engineering and Geodesy, Military University of Technology, Warsaw, Poland 2Institute of Navigation, Military University of Aviation, Deblin, Poland *E-mail of corresponding author: [email protected]

Figure 1 Horizontal trajectory of the UAV flight

Figure 2 Vertical trajectory of the UAV flight

The polynomial model is expressed by equation [6]: f - is a determined linear coefficient, X - indicates the number of the next measurement epoch, Yc=+$$Xd2 Xe+ , (2) g - is a determined linear coefficient. The mathematical Equations (1), (2) and (3) are solved where: using the least squares method, as below [8]: Y- stands for the value of the Heading, Pitch or Roll rotation angle, Aq$ d -=lv c - is a determined linear coefficient, NA= T $ A X - indicates the number of the next measurement epoch, LA= T $ l , (4) d and e are determined linear coefficients. dqN= -1 $ L The logarithmic model is expressed by formula [7]: vv m0 = nk6-@ Yf=+$ log Xg, (3) where: A - matrix of partial derivatives, where: d q- designated parameters, Y- stands for the value of the Heading, Pitch or Roll rotation l - vector of observation, angle, v - residuals vector,

VOLUME 22 COMMUNICATIONS 2/2020 54 WIERZBICKI, KRASUSKI

Table 1 Coefficients for the Heading angle

Research method Value of the linear coefficients Value of m0 parameter b= 270.70o Linear regression m = 1.09o a = -0.01o 0 e= 272.16o o o 2nd degree polynomial d = -0.22 m0= 0,91 c= 0.01o g= 271.60o Logarithmic trend m = 1.04o f= -0.92o 0

Table 2 Coefficients for the Pitch angle

Research method Value of the linear coefficients Value of m0 parameter b= 4.02o Linear regression m = 0.98o a = 0.03o 0 e= 4.00o o o 2nd degree polynomial d = 0.03 m0= 0.98 c= -0.01o g= 3.45o Logarithmic trend m = 0.97o f= 1.03o 0

Table 3 Coefficients for the Roll angle

Research method Value of linear coefficients Value of m0 parameter b= -0.01o Linear regression m = 1.43o a = -0.01o 0 e= 0.38o o o 2nd degree polynomial d = -0.01 m0= 1.42 c= 0.01o g= 0.62o Logarithmic trend m = 1.41o f= -0.57o 0

N - matrix of a set of normal equations, recorded by the sensor Trimble UX-5 in a universal text L - vector of free terms, format “log” [9]. m0 - standard deviation of residuals, The first stage of the experiment was to determine n - number of measurements, linear coefficients from Equations (1) to (3) for each k - number of designated parameters. HPR angle. The sought parameters from Equations (1) The mathematical model from Equation (4) is used for to (3) were determined based on the readings from 40 numerical calculations separately for each test method and measurement epochs (epochs 1 to 40). The values of linear independently of each HPR angle. coefficients from Equations (1) to (3) are shown in Tables 1, 2 and 3. Furthermore, those Tables present the value of

standard deviation of m0 for each HPR angle and each test

3 Research test and results method. In all the cases, values of the m0 parameter do not exceed 20, which constitutes a boundary value of accuracy The scientific experiment was carried out for real data for the HPR angles from the metric of the Trimble UX-5 obtained from the Trimble 5-UX device. For the experiment, instrument. The largest values of the m0 parameter occur the authors selected 80 exemplary measurement periods, for the Roll angle, being equal to approximately 1.50 and the for which the sensor Trimble UX-5-determined the position, lowest for the Pitch angle, being equal to approximately 10. flight altitude and the HPR values. Figures 1 and 2 illustrate the horizontal coordinates and the UAV flight altitude, respectively. The UAV coordinates were determined by 4 Discussion using the C/A code observations on the L1 frequency in the GPS system and were updated every 10 Hz. The UAV Within the discussion, the authors focused on the coordinates are referenced to the geodetic BLh frame and problem of extrapolation of the HPR angles’ values. To

COMMUNICATIONS 2/2020 VOLUME 22 METHODS OF PREDICTING THE HEADING, PITCH AND ROLL ANGLES FOR AN UNMANNED AERIAL... 55

Figure 3 Comparison of predicted and real values of the Heading angle in each measurement epoch this end, prediction of values of the HPR angles for the The predicted values of the HPR angles are compared measurement epochs 41-80 was made. New values of the to the real readings of the orientation angles from the sensor HPR angles for the prediction phase were specified based Trimble UX-5 on the basis of the following dependence [9]: on the functional relationships. dH =-HHpred real dH2 • the linear regression method, mH = nr6 -@1 dP =-PPpred real 2 Yapred =+$ Xbpred , (5) dP , (8) p = m nr6 -@1 dR =-RRpred real where: dR2 Y- stands for the extrapolated value of the angle of rotation R = m nr6 -@1 Heading, Pitch or Roll for epochs 41 – 80, a - linear coefficient determined for epochs 1- 40, where:

Xpred - number of measurement epoch from 41 to 80, dH - difference between the extrapolated value from a given b - linear coefficient determined for epochs 1- 40; test method and the real value of the Heading angle from the 2nd degree polynomial: the Trimble UX-5 sensor,

Rpred - extrapolated value of the Heading angle based on 2 Ycpred =+$$Xdpred Xepred + , (6) Equations (5), (6) and (7),

Preal - recorded value of the Heading angle from the UAV where: device, Y- stands for the extrapolated value of the angle of rotation nr - number of measurements, nr = 40,

Heading, Pitch or Roll value for epochs 41 – 80, mH - the error associated with the extrapolation of results c - linear coefficient determined for epochs 1- 40, signifies the matching error of predicted results ofthe

Xpred - number of measurement epoch from 41 to 80, Heading angle with respect to the actual values from the d and e - linear coefficients determined for epochs 1- 40; sensor Trimble UX-5. • logarithmic trend model: dP - difference between the extrapolated value from a given test method and a real value of the Pitch angle from the

Yfpred =+$ log Xgpred , (7) Trimble UX-5 sensor,

Rpred - extrapolated value of the Pitch angle based on where: Equations (5), (6) and (7),

Y- stands for the extrapolated value of the angle of rotation Preal - recorded value of the Pitch angle from the UAV device,

Heading, Pitch or Roll value for epochs 41 – 80, mp - the error associated with the extrapolation of results f - linear coefficient determined for epochs 1- 40, signifies the matching error of predicted results of the Pitch

Xpred - number of measurement epoch from 41 to 80, angle with respect to the actual values from the sensor g - linear coefficient determined for epochs 1- 40. Trimble UX-5,

VOLUME 22 COMMUNICATIONS 2/2020 56 WIERZBICKI, KRASUSKI

Figure 4 The comparison of predicted and real values of the Pitch angle in each measurement epoch

Figure 5 Comparison of predicted and real values of the Roll angle in each measurement epoch dR - difference between the extrapolated value from a given under 0.01o, with the scatter of results ranging from -2.37o test method and the real value of the Heading angle from to 1.99o. Moreover, the median value for parameter dH is the Trimble UX-5 sensor, 0.02o and the error of extrapolation is equal to 1.01o. In the

Rpred - extrapolated value of the Roll angle based on polynomial model, the average value of the dH parameter Equations (5), (6) and (7), equals approximately 8.11o, with the scatter of results o o Preal - recorded value of the Roll angle from the UAV device, ranging from 0.87 to 17.38 . Besides, the median value for o mR - the error associated with the extrapolation of results parameter dH in this model equals 7.28 and the error of signifies the matching error of predicted results of the Roll extrapolation is equal to 4.96o. The average value of the angle with respect to the actual values from the sensor parameter dH in the logarithmic method is equal to -0.13o Trimble UX-5. for the scatter of results ranging from -2.51o do 1.88o. The Figure 3 shows the dH parameter values for the Heading median value for the parameter dH in this model is equal to o o angle based on Equation (8). Values of the dH parameter -0.14 , while the extrapolation error for mH is equal to 1.01 . were determined for the linear regression method, the Based on obtained results, it can be concluded that results polynomial method and the logarithmic trend. In the linear of the parameter dH for the linear regression model and the regression model, the average value of the dH parameter is logarithmic trend definitely differ from the other ones.

COMMUNICATIONS 2/2020 VOLUME 22 METHODS OF PREDICTING THE HEADING, PITCH AND ROLL ANGLES FOR AN UNMANNED AERIAL... 57

Figure 4 shows the dP parameter values for the Pitch to reference results, the accuracy of Roll angle equals to 2o angle based on Equation (8). Values of the dP parameter and for Pitch angle more than 8o, respectively. In addition, were determined for the linear regression method, the in paper [13], the MPC method was used in Ariel UAV model polynomial method and the logarithmic trend. In the linear in Matlab Simulink environment. The algorithm is focused regression model, the average value of the dP parameter on differential equation for controlling the UAV dynamics. equals 1.34o, with the scatter of results ranging from -0.32o The paper presents good convergence between predicted to 3.99o. Moreover, the median value for the parameter dP is and real time data, with difference less than 0.5o for the HPR 1.20o and the error of extrapolation is equal to 1.13o. In the angles. In next article [14], the authors presented results polynomial model, the average value of the dP parameter of research for determination of the HPR angles in the equals approximately 1.25o, with the scatter of results prediction model. In that case, the Extended Kalman Filters ranging from -0.38o to 3.80o. Besides, the median value for based on GPS/INS fusion data was applied for prediction the dP parameter in this model equals 1.12o and the error the HPR angles for UAV. The accuracy of presented method of extrapolation is equal to 1.11o. The average value of the is about ±2o. In paper [15], the authors presented different parameter dP in the logarithmic method is equal to 0.64o scenarios for simulating the HPR angles, especially, the for the scatter of results ranging from -1.04o do 3.48o. The data from gyroscope model, accelerometer model, as well median value for the dP parameter in this model is equal to as magnetometers were utilized in simulation of the HPR o o 0.47 , while the extrapolation error for mP is equal to 1.08 . angles. The maximum orientation error of the Heading Based on obtained results, it can be concluded that results angle has reached up to ±30o, whereas for Pitch and Roll it of the dP parameter for the linear regression model and the was close to ±10o. polynomial model are similar. The largest scatter of the dP The obtained results of comparison of HPR predicted results and the error of extrapolation are noticeable in the values are better than results in papers [10] and [15]. In logarithmic model. addition, the simulated values of the HPR angles (e.g. Figure 5 shows the dR parameter values for the Roll especially for the Roll and Pitch angles) are close to results angle based on Equation (8). Values of the dR parameter included in papers [11] and [14]. Moreover, the obtained were determined for the linear regression method, the values of standard deviation of the HPR angles are lower polynomial method and the logarithmic trend. In the linear than values in paper [12]. Only results from the MPC regression model, the average value of the dR parameter method from paper [13] are better than presented values of equals -0.17o, with the scatter of results ranging from -3.40o the HPR angles in this article. Finally, the presented method to 2.18o. Moreover, the median value for the dR parameter of estimation of the HPR angles is a good and correct is -0.54o and the error of extrapolation are equal to 1.36o. solution for prediction model of the UAV orientation. In the polynomial model, the average value of the dR parameter equals approximately 1.99o, with the scatter of results ranging from -0.90o to 5.77o. Besides, the median 5 Conclusions value for the dR parameter in this model equals 2.07o and the error of extrapolation is equal to 1.69o. The average In paper, the scientific problem of comparison between value of the dR parameter in the logarithmic method is simulated and real data of the HPR angles of a UAV was equal to -0,41o for the scatter of results ranging from -3.66o presented. Especially the scientific problem should be do 1.99o. The median value for the dR parameter in this developed when the UAV object or navigational data model is equal to -0.77o, while the extrapolation error for from the UAV from flight mission would be lost. For that o mR is equal to 1.37 . Based on obtained results, it can be reason three scientific methods for recovery of the UAV concluded that the results of the dR parameter for the linear orientation were applied in this paper, i.e. the method of regression model and the logarithmic trend remain on linear regression, the model of the 2nd degree polynomial a similar level. The largest scatter of dR results and the error and the logarithmic trend. of extrapolation is noticeable in the polynomial model. For each test method, the values of the HPR angles In References there are many different algorithms were determined at the prediction stage and compared to for solving the problem of predication of the HPR angles the actual readings of the rotation angles from the device values. For example, in paper [10] the author proposed to Trimble UX-5. The experimental test was performed for prediction of HPR angles in the two mathematical functions: exemplary results of the HPR rotation angles from 80 polynomial and trigonometric. The maximum range measurement epochs. Based on the obtained findings, the between the real and predicted data of the HPR angles is following conclusions were formulated for accuracy of close to ±23o. In another paper [11], the authors present the each research method: Newton Euler’s mathematical model for controlled the UAV 1) the standard deviation term of linear regression method position. The maximum difference between the real and for the HPR angles is less than 1.4o; model data of the HPR angles reaches up ±2-3o. In the next 2) the standard deviation of the 2nd degree polynomial paper [12], the authors described the MPC (Model Predictive method for the HPR angles is less than 5o; Control) algorithm for quadrotor Unmanned Aerial Vehicle 3) the standard deviation of logarithmic method for the motion. In the research test, the Roll and Pitch angles were HPR angles is less than 1.4o. estimated in computer simulation. Based on comparison

VOLUME 22 COMMUNICATIONS 2/2020 58 WIERZBICKI, KRASUSKI

Finally, the presented research methods can be Acknowledgements developed and applied in technology of UAV in aerial navigation. The obtained results of the HPR angles’ This paper was supported by Military University of prediction are acceptable and suitable for the UAV Aviation and by Military University of Technology, Faculty technology. The typical accuracy of measurement of the of Civil Engineering and Geodesy, Institute of Geospatial o HPR angles equals to mHPR=2 [10]. Moreover, the obtained Engineering and Geodesy. results of accuracy of the HPR angles’ prediction are less than boundary error of the UAV orientation error, i.e. o 3·mHPR=6 . based on that, the presented research methods can be implemented in the UAV technology.

References

[1] IZVOLTOVA, J., PISCA, P., CERNOTA, P., MANCOVIC, M. Adjustment of Code Ranging of GNSS Observations. Communications - Scientific Letters of the University of Zilina [online]. 2016, 18(4), p. 15-18. ISSN 1335-420, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/283 [2] KEDZIERSKI, M., FRYSKOWSKA, A., WIERZBICKI, D. Opracowania fotogrametryczne z niskiego pulapu / Photogrammetric studies from low altitude (in Polish). Warsaw: Wydawnictwo Wojskowa Akademia Techniczna, 2014. ISBN 978-83-7938-047-3 [3] JAFERNIK, H., KRASUSKI, K., MICHTA, J. Assessment of suitability of radionavigation devices used in air. Scientific Journal of Silesian University of Technology. Series Transport [online]. 2016, 90, p. 99-112. ISSN 0209-3324, eISSN 2450-1549. Available from: https://doi.org/10.20858/sjsutst.2016.90.9 [4] GRYGIEL, R., BIEDA, R., WOJCIECHOWSKI, K. Metody wyznaczania katow z zyroskopow dla filtru komplementarnego na potrzeby okreslania orientacji IMU / Angles from gyroscope to complementary filter in IMU (in Polish). Przeglad Elektrotechniczny [online]. 2014, 90(9), p. 217-224. ISSN 0033-209, eISSN 2449-9544. Available from: https://doi. org/10.12915/pe.2014.09.52 [5] SCHNEIDER, A., HOMMEL, G., BLETTNER, M. Linear regression analysis: part 14 of a series on evaluation of scientific publications. Deutsches Arzteblatt International [online]. 2010, 107(44), p. 776-782. eISSN 1866-0452. Available from: https://doi.org/10.3238/arztebl.2010.0776 [6] CHEN, Y., XIE Z. Research on the trajectory model for ZY-3. The Scientific World Journal[online]. 2014, ID 429041. ISSN 2356-6140, eISSN 1537-744X. Available from: https://doi.org/10.1155/2014/429041 [7] SALAS, A. H. The logarithmic function as a limit. Applied Mathematical Sciences. 2012, 6(91), p. 4511 - 4518. ISSN 1312-885X, eISSN 1314-7552. [8] OSADA, E. Geodezja / Geodesy (in Polish). Wroclaw: Oficyna Wydawnicza Politechniki Wroclawskiej, 2001. ISBN 83-7085-663-2. [9] WIERZBICKI, D., KRASUSKI, K. Estimation of rotation angles based on GPS data from UX5 platform. Measurement Automation Monitoring. 2015, 61(11), p. 516-520. ISSN 2450-2855. [10] WIERZBICKI, D. The prediction of position and orientation parameters of UAV for video imaging. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences [online]. 2017, XLII-2/W6, p. 407-413. eISSN 2194-9034. Available from: https://doi.org/10.5194/isprs-archives-XLII-2-W6-407-2017 [11] ZOUAOUI, S., MOHAMED, E., KOUIDER, B. Easy tracking of UAV using PID controller. Periodica Polytechnica Transportation Engineering [online]. 2019, 47(3), p. 171-177. ISSN 0303-7800, eISSN 1587-3811. Available from: https:// doi.org/10.3311/PPtr.10838 [12] GRUJIC, I., NILSSON, R. Model-based development and evaluation of control for complex multi-domain systems: attitude control for a quadrotor UAV. Technical report ECE-TR-23. Aarhus, Denmark: Aarhus University, Department of Engineering, 2016. [13] RAEMAEKERS, A. J. M. Design of a model predictive controller to control UAVs [online]. DCT rapporten; vol. 2007.141. Eindhoven: Technische Universiteit Eindhoven, 2007. Available from: https://pure.tue.nl/ws/portalfiles/ portal/4337910/657983.pdf [14] LOPES, H. J. D. Attitude determination of highly dynamic fixed-wing UAVs [online]. Msc Thesis. Lisbon, Portugal: Instituto Superior T´ecnico, 2011. Available from: https://fenix.tecnico.ulisboa.pt/downloadFile/395143092026/ dissertacao.pdf [15] KOK, M., HOL, J. D., SCHON T. B. Using inertial sensors for position and orientation estimation [online]. In: Foundations and trends in signal processing. 2017, 11(1-2), p. 1-153. ISBN 9781680833560. Available from: http:// dx.doi.org/10.1561/2000000094

COMMUNICATIONS 2/2020 VOLUME 22 METHODS OF PREDICTING THE HEADING, PITCH AND ROLL ANGLES FOR AN UNMANNED AERIAL... 59

Annex - Nomenclature of used abbreviations

Abbreviation Full name GNSS Global Navigation Satellite System UAV Unmanned Aerial Vehicle INS Inertial Navigation System HPR Heading, Pitch and Roll IMU Inertial Measurement Unit GPS Global Positioning System C/A Code Acquisition BLh Latitude, Longitude, Ellipsoidal Height MPC Model Predictive Control

VOLUME 22 COMMUNICATIONS 2/2020 60 https://doi.org/10.26552/com.C.2020.2.60-67 REVIEW ARTICLE

Rafal S. Jurecki ANALYSIS OF ROAD SAFETY IN POLAND AFTER ACCESSION TO THE EUROPEAN UNION

The article presents an assessment of the state of safety on Polish roads in the period 2004-2018 in terms of changes that occurred after Poland’s accession to the EU. The number of vehicles in the period in question was analyzed and the annual growth of the number of vehicles of the main types was assessed. Analysis of changes in the length of the most important road types was carried out. The number of accidents and fatalities was analyzed and the changes in this area were assessed. The statistical decrease in the number of accidents and fatalities, as well as a very strong correlation between them, were confirmed. It was indicated in which groups of accidents the fatality risk is the highest. The paper presents a summary of the types of roads on which traffic accidents and fatal traffic accidents occurred. Shares of these roads were determined, indicating that the most dangerous roads in Poland are the two-way, two-lane roads. Keywords: road safety, accident, accident statistics

1 Introduction road factors [13-14], etc. Problems related to road safety were analyzed in a number of publications [15-17]. The use of cars has become commonplace in everyday In 2004, Poland was included into the European Union. life. Unfortunately, the road traffic is a source of traffic Since then, significant changes have taken place in many events hazardous to the health and life of road users. Every aspects of life in Poland. The possibility of free movement year, there are tens of thousands of accidents in Poland, in Europe within the Schengen Area, liberalization of with several thousand fatalities. Tens of thousands of many regulations caused changes in the structure of people are injured in these accidents, which is of no less exploited vehicles caused by a large influx of used vehicles. importance. Traffic accidents have significant social and At the same time, along with numerous investments in economic consequences [1-2]. In 2004, the unit cost of infrastructure, significant changes have taken place in the a fatality in Poland, estimated according to the PANDORA broadly understood road surroundings. 2014 method, was approximately PLN 2.05 million and that The article analyzes selected factors that impact the of severely injured person was PLN 2.32 million. In Poland, safety on Polish roads. the costs of accidents in the year 2004 amounted to about PLN 30 billion [2]. Due to the significance of the traffic accidents problem, 2 Vehicles in Poland a number of publications [3-5] have undertaken analysis of the state of safety on roads of a multitude countries, The number of vehicles registered in Poland in the including Poland [6-8]. This is despite the fact that the analyzed period has significantly changed. In nearly 15 roads in the EU are considered to be among the safest in years, it has increased from approximately 16.7 million to the world. Among those involved in the matters of the road nearly 30 million. If we take the 2004 (as 100%) as a baseline safety, one question is still being raised: what solutions value, the recorded increase in this respect reached 77% should be implemented to reduce accident rates and their [18]. negative consequences? In order to consider these issues, At the same time, the number of passenger cars has an in-depth analysis of the causes and consequences of increased from nearly 12 million to over 22.5. In turn, this traffic accidents is necessary, taking into account various translates into an increase of nearly 90%. The number factors [9]. Factors that may determine the possibility of of trucks (44%) and motorcycles (71%) also increased. traffic accidents include: the state of road infrastructure in A summary of the numbers of motor vehicles and passenger a given country [10], technical factors [11], cultural factors, cars is presented in Figure 1. When analyzing the number of e.g. inclination to respect the provisions of the Road registered vehicles and population [19], it can be observed Code, alcohol consumption, psychological factors [12], that the number of passenger cars per 1000 residents in

Rafal S. Jurecki Department of Automotive Engineering and Transport, Faculty of Mechatronics and Automotive Engineering, Kielce University of Technology, Poland E-mail of corresponding author: [email protected]

35 3.5 15 Total Passengers Cars Total Trucks Motorbikes Passengers Cars 30 3 Trucks 10 Motorbikes , milion 25 +77.6% 2.5 5 20 2

Motorbikes 0 milion 15 1.5 and and 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 10 1 -5 Trucks

5 0.5

Increaseof Number of vehicles, % -10 umber of of umber N Total number of vehicles and Passengers Cars, Cars, Passengers and of vehicles number Total 0 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 -15 Year Year

Figure 1 Numbers of vehicles by type in Poland between Figure 3 Relative change in numbers of vehicles 2004 and 2018

2300 2500 Total Passengers Cars Trucks 1800 Motorbikes 2000 +712%

1300 1500

+206% 1000 800

Length of roads, km roads, of Length motorways 100% expressways 500 300 100%

Increase of Number of vehicles, thousand vehicles, of Number of Increase 0 -200 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Year Year Figure 2 Change in the numbers vehicles by type Figure 4 Change in the length of the most important roads in Poland in the years 2004-2018

Poland increased form 314 in 2004, and to 586 cars in by slightly more than 10%. However, the changes are 2018. In this respect, Poland is currently one of the leading particularly visible in the case of national roads [21], as well European countries. as expressways and motorways [22]. After the borders were opened and procedures such as The change in the length of the most important roads customs formalities were simplified, the annual increases in in Poland is presented in Figure 4 [22]. At the end of 2018, the number of cars were significant, as shown in Figure 2. the total length of expressways and motorways in Poland One of the main factors influencing the increase in amounted to 2092 km and 1638 km, respectively. In total, the total number of vehicles was the change in the number these roads in relation to all the public roads in Poland of passenger cars. In the record-breaking year of 2008, constitute only just over 0.8% of all roads. the growth of all registered vehicles reached 1.85 million In Europe, Spain has the most developed motorway vehicles, 1.5 million of which were passenger cars. If the network, having more than 17 thousand km of motorways relative changes in the number of vehicles (Figure 3) in on an area of 500 thousand km2 [6]. In Germany (total particular groups of vehicles are analyzed, it is clear that area of 357 thousand km2) there are 15.3 thousand km of the only decrease was recorded in 2005 (compared to 2004) such roads, while in France (551 thousand km2) -just over with regard to the number of trucks and motorcycles. 11.6 thousand km of such roads [18, 23]. The length of In 2008, the highest relative increase in the number such roads per total area of a country is an indicator that of passenger cars, motorcycles and trucks was recorded determines the degree of development of the road network -7-10% [18]. and the density of the best roads. In Poland, it is only 0.005 km/1 km2, while in Belgium it is 0.058 km/1 km2, Holland 0.074 km/1 km2, Spain 0.0344 km/1 km2. 3 Changes to roads in Poland Apart from expressways and motorways [22], Poland has nearly 20 thousand km of national roads [24]. All of There are over 420 thousand public roads in Poland [8, them include the most important road routes managed 20]. In recent years, especially after Poland’s accession to by the Polish General Directorate for National Roads and the EU, the road network has undergone certain changes. Motorways (GDDKiA). In Poland, these constitute nearly The length of public roads has increased relatively little, 5% of all available roads.

VOLUME 22 COMMUNICATIONS 2/2020 62 JURECKI

8 70 000 6 Relative Change of Number of Accidents 2004 60 000 4 2

50 000 % Accidents,

of 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 40 000 -2

-4 30 000 Mean value -3.24% -6

Number of Accidents of Number 20 000 -8

10 000 -10

Relative Change of Number of Change Relative -12 0 -14 Year, compared to the previous one 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 Year Figure 5 Number of traffic accidents in Poland Figure 7 Relative change in the number of accidents

6 000 100 % 60 000 100% 5 000 50 000 -49.9% 4 000 40 000 -38%

3 000 30 000

umber of fatalities 2 000 20 000 N Number of Accidents of Number

10 000 1 000

0 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Year Year Figure 6 Decrease of number of traffic accidents in Poland Figure 8 Number of fatalities on Polish roads in 2004-2018 in 2004-2018

If 2004 is assumed as the baseline for the state of among the top European countries. In this respect Poland roads, in nearly fifteen years the length of motorways was surpassed by such countries as: Germany with 339 has increased by more than 206%, while the length of thousand accidents, Italy -243 thousand, Great Britain -213 expressways has increased by 712%. thousand or, the similar in terms of area and population, Spain -94 thousand. It should be noted, however, that the way in which accidents are counted and defined varies 4 Traffic accidents, their causes and consequences from country to country, making it very difficult to compare these data in full. During a meeting of EU transport The road traffic generates the danger of collisions ministers in Valetta, attention was also drawn to the fact and accidents. Unfortunately, these incidents are still an that EU Member States should report traffic accidents using integral part of the road traffic. Despite many actions taken a common definition of “accidents”, using comparable and both in Poland [25-27] and by EU authorities [28-30], the reliable data [28, 34]. still unattainable and still time-shifted goal is to halve the Since 1997, there has been a clear decrease in the number of deaths. Such an ambitious goal was set earlier number of accidents in Poland. When the analysis is for the year 2020 (in relation to 2010) [28, 30]. narrowed down to the period 2004-2018 (see Figure 6) it In Poland, in the analyzed years, considering the can be seen that the number of accidents since Poland’s significant density of traffic, the statistically assessed road accession to the EU has decreased by 38%. It is worth safety improved, although it was subject to significant noting, however, that at the same time in the analyzed fluctuations [7, 16, 31]. To make the state of safety in Poland period the number of vehicles increased by nearly 80%. easier to assess, Figure 5 shows the number of accidents in The value of the relative change in the number of the last 40 years. accidents is shown in Figure 7, the average value (marked From such a perspective, it is clear that the culmination with a red line) in the analyzed years amounted to -3.24%. point of the number of accidents on Polish roads (which This indicates an annual statistical decrease in the number occurred many years before Poland’s accession to the of accidents. The largest decrease, i.e. approx. 10% and EU) was in 1997. A record number of over 66 thousand over, was recorded in the years 2009-2010. However, the accidents was recorded in Poland in that year. In order downward trend was not recorded annually, as in 2007 the to indicate the importance of the problem, it is worth highest increase in the number of accidents was recorded, mentioning that according to EC (CARE) data [23, 32-33] reaching 6%. there were more than 1.36 million accidents in Europe total. In 2004, a total of 47,900 people died in EU countries. In In terms of the number of accidents, this placed Poland Poland, this value was 5,712. Calculated per million citizens,

COMMUNICATIONS 2/2020 VOLUME 22 ANALYSIS OF ROAD SAFETY IN POLAND AFTER ACCESSION TO THE EUROPEAN UNION 63

Table 1 Coefficient of determination 2R for the accident data in different European countries Country Coefficient of determination R2 Country Coefficient of determination 2R Austria 0.450 Italy 0.912 Belgium 0.754 Latvia 0.893 Bulgaria 0.733 Lithuania 0.969 Croatia 0.934 Luxembourg 0.870 Czech Republic 0.828 Malta 0.043 Cyprus 0.830 Netherlands 0.714 Denmark 0.827 Poland 0.994 Estonia 0.981 Portugal 0.819 France 0.941 Romania 0.006 Finland 0.754 Slovakia 0.667 Germany 0.922 Slovenia 0.919 Greece 0.928 Spain 0.230 Hungary 0.982 Sweden 0.960 Ireland 0.249 UK 0.960 EU 0.988

7000

6000 R² = 0.9405 R² = 0.9119 R² = 0.9943 Poland 5000 R² = 0.9215 Belgium Denmark

of fatalities 4000 France R² = 0.9598 R² = 0.1837 Great Britain 3000 Spain Number R² = 0.0029 Hungary 2000 Germany R² = 0.9824 Italy 1000 Romania R² = 0.7542 R² = 0.667 Slovakia 0 R² = 0.8272 0 50000 100000 150000 200000 250000 300000 350000 400000 Number of acidents Figure 9 Number of road accidents versus the number of fatalities for several member states of the EU this means that, in 2004, there were almost 150 fatalities in Analysing the available information presented by CARE traffic accidents per million citizens. In the period 2004-2018 [23, 33], one wonders whether there is any relationship the number of fatalities on Polish roads decreased by nearly between the number of road accidents and the number of 50% (Figure 8). fatalities. From the data presented in [34] it is clear that In 2018, 3,862 fatalities were recorded, which means the number of accidents in Poland is largely dependent on that statistically 76 people per 1 million citizens died on the number of people killed. If one compares the results the roads in Poland. Taking into account the fact that the provided in Figure 9, it can be seen that the coefficient EU average in 2018 in this respect was 49, Poland ranked of determination R2 for Poland is very high, reaching fifth in this unfavorable ranking. Poland was ahead of approximately 0.99. High values are also reported for such Romania (96), Bulgaria (88), Latvia (78) and Croatia (77) in countries as Hungary (R2= 0.982), Estonia (R2= 0.981), this respect. United Kingdom (28), Denmark (29), Ireland Lithuania (R2= 0.982) and the UK (R2= 0.96). The values and Netherlands (31) and Sweden (32) have been identified indicate that if the number of road accidents is reduced, the as the safest countries in this respect [33]. According to number of fatalities will decrease, too. However, reducing incomplete data, in 2018 there were around 25,300 fatalities the number of killed to zero in this way is not possible in and 135,000 injured on the EU roads [30]. practice. The mean coefficient of determination obtained

VOLUME 22 COMMUNICATIONS 2/2020 64 JURECKI

10 2 500 Relative Change of Fatilities 5 2 000 R² = 0.9743 side collision

0 1 500 head-on collision 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 rear-end collision R² = 0.9812 -5 1 000 Mean value -4.55% pedestrian collision R² = 0.9775 of fatalities in accidents in of fatalities R² = 0.834 collision with a tree -10 500 rollover of the vehicle number -15 R² = 0.6658 R² = 0.7146 0 Relative Change in Accidents, % of Fatilities 0 5 000 10 000 15 000 20 000 -20 Year , compared to the previous one number of accidents

Figure 10 Relative decrease in the number of fatalities Figure 13 Dependency of number of fatalities on number of accidents

35 100 side collision % 30 head-on collision 95

25 rear-end collision 90

20 pedestrian collision 85 15 collision with a tree

rollover of the vehicle 80 10 relative share of accidents, % accidents, of share relative 75 5

year roadsof types on different accidents of Share 70 0 2004 2006 2008 2010 2012 2014 2016 2018 Year Figure 11 Specificity of traffic accidents in Poland Figure 14 Share of accidents on different types of roads in Poland 40

35 100 side collision % 30 95 head-on collision

25 rear-end collision 90

20 pedestrian collision 85

15 collision with a tree on different types of roads 80 rollover of the

relative sharefatalities,of 10 %

vehicle fatalities 75 5 year Share of 70 0 2004 2006 2008 2010 2012 2014 2016 2018 Year

Figure 12 Specificity of fatalities in the main types of traffic Figure 15 Graphic representation of the shares of fatalities accidents in Poland on particular types of roads in Poland for the total number of accidents and the total number of 7%). The average decrease in the relative number of road fatalities in the EU member states is R2=0.988. The lowest fatalities in the analyzed period was 4.55%. values of R2 are observed for Romania (R2= 0.006), Malta When analyzing the state of the road safety in the (R2= 0.043) and Luxembourg (R2= 0.087), which suggests period in question, accidents where divided in the main that there is no such relationship for these countries. types as presented in Figure 11. Table 1 shows the values of the coefficient of Analyzing the specificity of accidents, it is worth noting determination R2, being the ratio of the number of road that in the years 2004-2019, the share of side collisions accidents to the number of fatalities, obtained for selected caused by e.g. not yielding when appropriate, as well as EU member states in the analysed period. rear-end collisions and rollovers, significantly increased. Figure 10 presents the results of analysis of the relative It is worth emphasizing that, in the analyzed period, the annual changes in the number of traffic accident fatalities. share of pedestrian collisions significantly decreased from In certain years (especially in 2009, 2010 and 2012) 33% to about 24% [18, 35]. Despite the number of accidents quite a significant reduction of their value -by about 15%, involving pedestrians still being high, this decrease can be was recorded. Only in four years of the analyzed period, deemed positive, as almost every accident with participation an increase was recorded, with the highest in 2011 (by

COMMUNICATIONS 2/2020 VOLUME 22 ANALYSIS OF ROAD SAFETY IN POLAND AFTER ACCESSION TO THE EUROPEAN UNION 65

of the so-called unprotected road users results in a fatality Comparing Figures 14 and 15, there are certain clear or serious injury [25]. trends in the analyzed period. The share of accidents on Figure 12 presents the share of fatalities in this type type “A” roads, i.e. two-way, two-lane roads is decreasing of accidents. The largest, although decreasing (over 30%) (a decrease of about 4% was recorded), but it did not result share in the number of fatalities are accidents related to in a significant decrease in the share of fatalities on these pedestrians being run over. This is followed by the head-on roads (about -0.6%). A minor increase (by about 1%) in collisions (20%), although it can be seen that the side the number of accidents on other types of roads resulted collisions have been equally dangerous in recent years. in a decrease in the share of fatalities on roads type “B” Figure 13 presents the results of the correlation by 2.3% and the increase on motorways (type “M”) and analysis between the number of accidents and the number expressways (type “E”) by slightly more than 1%. of fatalities in different types of accidents. When analyzing the graph in Figure 13, it can be seen that the high value of the coefficient of determination confirms 5 Conclusions the close dependency between the number of accidents In 2004, Poland has joined the European Union. Since and fatalities. The lowest coefficient of determination then, a visible qualitative change is taking place in many occurs in case of rollovers and amounts to R2=0.665. The areas in Poland. In the area of transport and road safety, highest coefficients were recorded for head-on collisions certain changes and regularities could also be observed (R2=0.981), collisions with trees (R2=0.977) or pedestrian over the past 15 years. It is easily noticeable that the collisions (R2=0.971), which means that these types of number of registered cars in Poland in 2018 grew to nearly accidents have the greatest impact on the number of 30 million with over 22.5 million passenger cars. When fatalities. analyzing these data in relation to the base year 2004, one Figure 14 presents a summary of accidents occurring can speak of increases of 77.6% and 88%, respectively. on different types of roads. The following types of roads The annual increase of the number of vehicles in certain were considered: years was as high as 10%. These vehicles make use of a) “A” -two-way, two-lane roads, the available road network. The total length of all public b) “B” -one-way, two-lane roads, roads in Poland increased by about 10%, in the case of the c) “C” -one-way, single-lane roads, most significant roads and best in terms of their technical d) “E” -expressways, characteristics, the increase was more pronounced. The e) “M” -motorways. changes in the linear infrastructure were accompanied by The dominant type of roads (and at the same time changes in the statistics of traffic incidents. In the analyzed the most common in Poland) with regards to number of period, the number of accidents dropped by 38% and the accidents recorded were type “A” roads. Despite a number number of fatalities by nearly 50%. of investments, the share of accidents on these roads was The high values of the coefficient of determination 2R over 80%. On type “B” roads the percentage of accidents reported for several European countries, including Poland, was approx. 13%, on type “C” roads it was around 3-4%, confirm that one of the ways to reduce the number of while on roads type “M” and “E” it was approx. 1.5% [18]. fatalities is reducing the number of accidents. Since the share of accidents on roads type “C”, “E” and Analyzing the number of accidents and fatalities, “M” is relatively small, the range of values on the ordinate a very strong correlation (R2>0.97) between the number axis in the graph presented in Figure 14 was narrowed of accidents and the number of fatalities was confirmed, down to 70%-100% in order to improve its legibility [18]. in case of certain types of accidents: head-on collisions, Figure 15 presents a summary of the types of roads collisions with trees or pedestrians. Analyzing the structure on which fatal accidents are recorded. The dominant type of the road network, a summary was made of the types of of roads in this respect were type “A” roads, i.e. two-way, roads on which accidents and fatal accidents took place. two-lane roads. The share of fatalities on these roads was The most dangerous roads in Poland are two-way, two-lane nearly 86-88%. roads.

References

[1] JAZDZIK-OSMOLSKA, A. Wycena kosztow wypadkow i kolizji drogowych na sieci drog w Polsce na koniec roku 2015 - z wyodrebnieniem srednich kosztow spoleczno-ekonomicznych wypadkow na transeuropejskiej sieci transportowej, IBDiM / Valuation of the costs of road accidents and collisions on the road network in Poland at the end of 2015 - with isolation of the average socio-economic costs of accidents on the trans-European transport network (in Polish) [online]. Warszawa: Wydawca Ministerstwo Infrastruktury i Rozwoju, Sekretariat KRBRD, 2016. Available from: http://www.krbrd.gov.pl/files/file_add/download/393_wycena-kosztow-wypadkow-i-kolizji-drogowych-2015.pdf [2] WESZCZAK, A. (2018). Ekonomiczne i spoleczne determinanty bezpieczenstwa ruchu drogowego w Polsce. Analizy regionalne / Economic and social determinants of road safety in Poland. Regional analyzes (in Polish).

VOLUME 22 COMMUNICATIONS 2/2020 66 JURECKI

PhD dissertation [online]. Available from: http://dspace.uni.lodz.pl/xmlui/bitstream/handle/11089/25186/BRD. pdf?sequence=1&isAllowed=y [3] CABAN, J., DROZDZIEL, P., LISCAK, S. Wybrane aspekty stanu bezpieczenstwa w transporcie drogowym / Selected aspects of the state of safety in road transport (in Polish). Zeszyty Naukowe Instytutu Pojazdow. 2012, 3(89), p. 13-19. ISSN 1642-347X. [4] CABAN, J., DROZDZIEL, P., MARCZUK, A., KLUSKA, A. Przyczyny powstawania wypadkow drogowych w Polsce / Causes of road accidents in Poland (in Polish). Logistyka. 2015, 5, p. 703-710. ISSN 1231-5478. [5] GONIEWICZ, M., GONIEWICZ, K. Wypadki drogowe w Polsce - czynniki sprawcze i zapobieganie / Road accidents in Poland - the causative factors and prevention (in Polish). Bezpieczenstwo Pracy: Nauka i Praktyka. 2010, 09, p. 14-17. ISSN 0137-7043. [6] Hiszpania na pierwszym miejscu. Spain in the first place (in Polish) [online]. Available from: https://inzynieria.com/ wpis-branzy/rankingi/5/53404,w-ktorym-kraju-ue-jest-najdluzsza-siec-autostrad-ranking,pozycja-rankingu-hiszpania- na-pierwszym-miejscu [7] JASKIEWICZ, M., JURECKI, R., LISCAK, S. STOKLOSA, J. Analysis of the state of road safety in Swietokrzyskie voivodeship. XIV International Technical Systems Degradation Conference: proceedings. 2015, p. 25-28. [8] JURECKI, R., POLIAK, M. Traffic safety of main roads in Poland. New Trends in Production Engineering [online]. 2018, 1(1), p. 63-72. ISSN 2545-2843. Available from: https://doi.org/10.2478/ntpe-2018-0008 [9] JASKIEWICZ, M., JURECKI, R. The analysis of safety on Polish roads between 2000 and 2010. Transport and Communication. 2013, 1, p. 8-40. ISSN 1339-5130. [10] BARCIK, J., CZECH, P. Wplyw infrastruktury drogowej na bezpieczenstwo ruchu-czesc I / Condition of rail transport in Poland in recent years - part 1 (in Polish). Zeszyty Naukowe Politechniki Slaskiej. Seria Transport / Scientific Journal of Silesian University of Technology. Series Transport [online]. 2010, 67, p. 13-21. ISSN 0209-3324, eISSN 2450-1549. Available from: http://sjsutst.polsl.pl/archives/2010/vol67/005_ZN67_2010_BarcikCzech.pdf [11] WINTER, W., ZOLTOWSKI, B. Czynniki techniczne w ryzyku powstawania wypadkow drogowych / Technical factors in the risk of road accidents (in Polish). Zeszyty Naukowe Politechniki Swietokrzyskiej. Mechanika. 2006, 84, p. 339-350. ISSN 0239-4979. [12] BAK, J., BAK-GAJDA, D. Psychologiczne czynniki bezpieczenstwa ruchu drogowego / Psychological factors of road safety (in Polish). Maintenance and Reliability - Eksploatacja i Niezawodnosc. 2008, 3(39), p. 22-29. ISSN 1507-2711. [13] MATUSZAK Z., NOWAK, A., ZABINSKA, I., MARTINEZ DE OSES, F. X. Road safety on the example of the city of Bytom. The Archives of Automotive Engineering - Archiwum Motoryzacji [online]. 2019, 83(1), p. 43-57. eISSN 2084-476X. Available from: https://doi.org/10.14669/AM.VOL83.ART3 [14] METELSKI, A. Analysis of selected methodological problems regarding the examination of traffic events at road intersections. The Archives of Automotive Engineering - Archiwum Motoryzacji [online]. 2018, 82(4), p. 75-85. eISSN 2084-476X. Available from: https://doi.org/10.14669/AM.VOL82.ART6 [15] BROZYNA, E. Czynnik ludzki a bezpieczenstwo w ruchu drogowym / The Human factor and road safety (in Polish). Autobusy: Technika, Eksploatacja, Systemy Transportowe. 2017, 7-8, p. 49-52. ISSN 1509-5878. e-ISSN 2450-7725. [16] KISILOWSKI, J., ZALEWSKI, J. Wybrane problemy analizy przyczyn wypadkow drogowych w Polsce w latach 1995- 2004 / Selected problems of analyzing the causes of road accidents in Poland in 1995-2004. Maintenace and Realibility - Eksploatacja i Niezawodnosc. 2008, 1, p. 37-43. ISSN 1507-2711. [17] RACZYNSKA-BULAWA, E. Bezpieczenstwo w ruchu drogowym w Europie: zalozenia polityki UE i ocena podejmowanych dzialan z perspektywy danych statystycznych / Traffic Safety in Europe - the policy of UE and evaluation activities undertaken from the perspective of statistical data. Autobusy: Technika, Eksploatacja, Systemy Transportowe. 2016, 10, p. 8-14. ISSN 1509-5878. e-ISSN 2450-7725. [18] Annual reports of Police HQ Accidents 2004-20018 (in Polish) [online]. Available from: http://statystyka.policja.pl/st/ ruch-drogowy/76562,Wypadki-drogowe-raporty-roczne.html [19] Ludnosc. Stan i struktura oraz ruch naturalny w przekroju terytorialnym w 2018 r. Stan w dniu 31 XII / Population. Condition and structure, as well as natural movement in territorial division in 2018. As of 31 XII (in Polish) [online]. Available from: http://stat.gov.pl/obszary-tematyczne/ludnosc/ludnosc/ludnosc-stan-i-struktura-oraz-ruch-naturalny-w- przekroju-terytorialnym-w-2018-r-stan-w-dniu-31-xii,6,25.html [20] Drogi publiczne/Public roads [online]. Available from: http://drogipubliczne.eu/drogi-publiczne-w-liczbach [21] National roads in Poland. (in Polish) [online]. Available from: https://www.gddkia.gov.pl/pl/927/drogi-krajowe [22] Motorways in Poland (in Polish) [online]. Available from: https://www.gddkia.gov.pl/pl/926/autostrady [23] CARE (EU Road Accidents Database) [online] Available from: https://ec.europa.eu/transport/road_safety/specialist/ statistics_en# [24] Program budowy drog krajowych / National road construction program (in Polish) [online]. Available from: https://www.gddkia.gov.pl/pl/4059/Program-Budowy-Drog-Krajowych-sprawdz-stan-zaawansowania

COMMUNICATIONS 2/2020 VOLUME 22 ANALYSIS OF ROAD SAFETY IN POLAND AFTER ACCESSION TO THE EUROPEAN UNION 67

[25] JURECKI, R., JASKIEWICZ, M., WIECKOWSKI, D. Bezpieczenstwo niechronionych uczestnikow ruchu drogowego w Polsce / Safety of unprotected road users in Poland (in Polish). Autobusy. Technika, Eksploatacja, Systemy Transportowe. 2016, 11, p. 75-81. ISSN 1509-5878. e-ISSN 2450-7725. [26] K.R.B.R. Narodowy Program Bezpieczenstwa Ruchu Drogowego 2013-2020. Material na posiedzenie plenarne KRBRD 20.06.2013 / National Road Safety Program 2013-2020. Material for the KRBRD plenary session on 20/06/2013 (in Polish) [online]. Available from: http://www.krbrd.gov.pl/files/file/NP-BRD-2020_przyjety_przez_KRBRD.pdf [27] PLACHECKA, M. Efekty dzialan na rzecz poprawy bezpieczenstwa w transporcie publicznym w Polsce / Effects of actions to improve safety in public transport in Poland (in Polish). Autobusy: Technika, Eksploatacja, Systemy Transportowe. 2016, 10, p. 20-23. ISSN 1509-5878. e-ISSN 2450-7725. [28] Declaration Transport ministers of the Member States of the European Union. Meeting in Valletta on 29 March 2017 [online. Available from: http://data.consilium.europa.eu/doc/document/ST-6273-2017-INIT/en/pdf [29] Road safety citizen [online]. Available from: https://ec.europa.eu/ transport/road_safety/sites/roadsafety/files/pdf/ road_safety_citizen/road_safety_citizen_100924_en.pdf [30] Statystyki umieralnosci na drogach w UE / Road mortality statistics in the EU (in Polish) [online]. Available from: http://www.europarl.europa.eu/news/pl/headlines/society/20190410STO36615/statystyki-smiertelnosci-na-drogach-w- ue-infografika [31] WILK, T., GAWRON A. Analiza stanu bezpieczenstwa ruchu drogowego dla obszaru miasta / Analysis of road traffic safety for the city area (in Polish). Logistyka. 2014, 3, p. 1698-1706. ISSN 1231-5478. [33] EU Road Safety Policy Framework 2021-2030 - Next steps towards “Vision Zero” [online]. Available from: https://ec.europa.eu/transport/road_safety/sites/roadsafety/files/move-2019-01178-01-00-en-tra-00_3.pdf [34] 2018 road safety statistics [online]. Available from: https://europa.eu/rapid/press-release_MEMO-19-1990_en.htm [34] Mobility and transport, road safety [online]. Available from: https://ec.europa.eu/transport/road_safety/specialist/ statistics_en# [35] JURECKI R. Wybrane zagadnienia bezpieczenstwa pieszych na drogach w Polsce / Selected issues of pedestrian safety on roads in Poland (in Polish). Technika Transportu Szynowego. 2015, 12, p. 746-749. ISSN 1232-3829, e-ISSN 2543-5728.

VOLUME 22 COMMUNICATIONS 2/2020 68 https://doi.org/10.26552/com.C.2020.2.68-72 ORIGINAL RESEARCH ARTICLE

Jan Steininger - Stefan Medvecky - Robert Kohar - Tomas Capak OPTIMIZATION PROCEDURE OF ROLLER ELEMENTS GEOMETRY WITH REGARD TO DURABILITY OF SPHERICAL ROLLER BEARINGS

The article deals with an optimization procedure of roller elements geometry with regard to durability of spherical roller bearings. The aim of the article is to examine the impact of change of the roller elements inner geometry on durability and reliability of spherical roller bearings; the contact strain along a spherical roller by means of the Finite Element Method at contact points of components of a spherical roller bearing by means of designed 3D parametric models. The most appropriate shape of roller elements inner geometry of a bearing from the standpoint of calculated durability was determined based on results of the contact analyses. Keywords: rolling element, contact strain, spherical roller bearing

1 Introduction Roller bearings durability depends on a revolution number, which the bearing can perform until fatigue of any Roller bearings are an inseparable part of most machines of their components takes place. A peeled material is a sign and devices in which there takes place a rotation movement of the component fatigue. Fatigue is a basic and natural way or a linear motion. There are different requirements on of bearing damage. It is demonstrated by the presence of roller bearings. Production machines need bearings, which small cracks under the bearing raceway surface. The depth are able to work in high revolution, in power engineering of these cracks is usually about 0.05-0.3 mm depending bearings have to carry heavy loads, etc. on the surface curve radii of rolling elements and the Development, or rather rolling bearing optimization, bearing ringsraceways. The crack depth allows the material is conditioned by an increase of technical parameters in changes, which are caused by the slide pulsating strain. machines and devices. This fact refers especially to an This process leads to a gradual crack formation under the increase of input parameters such as power and revolution, surface. It can take quite a long time until it is visible on the weight and volume reduction, noise level reduction, surface in a form of the peeled off material, the so-called etc. However, the most important parameters requiring pitting [2-3]. optimization are the bearing lifetime and reliability. Development of new technologies introduces also new construction materials, new production techniques 3 Contact strain along spherical roller of semi-finished products and bearing components or in a spherical roller bearing new installation methods. It is important not to overlook the bearing construction. Here it is possible to perform It is possible to calculate the intensity of the contact geometry adjustment optimization. This adjustment applies pressure and the size of the contact surface - effective length especially to geometry adjustment of raceways and rolling lef and width 2b from the contact pressure distribution at elements in the spherical roller bearings. the most strained point in the bearing inner ring. Figure 1 shows the course projection (the curve) of the contact

pressure along the contact surface lef of the contact ellipse 2 Spherical roller bearings durability on the bearing inner ring. The contact strain curve has been calculated using the finite element method [4]. The double-row angular spherical roller bearing has a raceway spherically ground on the outer ring. The bearing is able to accommodate very high radial loads, as well as 4 Optimization of geometry of spherical roller heavy axial loads in both directions. The high radial load bearing capacity is caused by the great number of rolling elements, the so-called spherical rollers and their close contact on the The bearing model was simplified by axial symmetry. inner ring raceways [1]. the bonds between the individual parts of the bearing were

Jan Steininger1, Stefan Medvecky1, Robert Kohar2,*, Tomas Capak2 1Institute of Competitiveness and Innovations, University of Zilina, Slovak Republic 2Department of Design and Mechanical Elements, Faculty of Mechanical Engineering, University of Zilina, Slovak Republic *E-mail of corresponding author: [email protected]

Figure 1 The contact pressure course on the inner bearing ring raceway

Figure 2 Comparison of the contact pressure courses in the bearing raceway of the spherical roller bearing inner ring for the rolling element analyzed geometries replaced by contacts. Degrees of freedom were taken from The aim of optimization was a decrease of contact the outer ring of the bearing. The rolling element load pressure that acts at the point of contact of the rolling calculation is based on the Hertz theory and the reference element with the outer and the inner ring. The profile of bearing calculation data. It is possible to distribute the a rolling element was optimized and the contact strain bearing load to individual rolling elements. The load on the between the rolling elements and bearing rings was selected part of the bearing was calculated to be 9 kN. The calculated, as well [7]. contact volumes were meshed with 0.08 mm hexahedrons. The three new geometries of the rolling element The other parts were meshed with 0.56 mm tetrahedrons. for spherical roller bearings were designed that were The transition edges were meshed at 0.24 mm [5-6]. consequently compared with the reference profile. Selection of the most appropriate design of the new geometry of the

VOLUME 22 COMMUNICATIONS 2/2020 70 STEININGER et al.

Figure 3 Comparison of the contact pressure courses in the bearing raceway of the spherical roller bearing outer ring for the rolling element analyzed geometries

Table 1 Evaluation of results of analyses in the spherical roller bearing inner ring

Design nr. Title lef (mm) bef (mm) po (MPa) σmax (MPa) Reference Profile 21 0.96 2106.1 1426 bearing Cylindrical surface at the point 1 22.14 1.21 4022.1 3019.1 of contact 2 The change of the main radius 22.13 0.96 1921.5 1319.9 3 The combination of 2 radii 22.14 0.64 2106.1 1315.3 4 Logarithmic curve 21.9 0.63 1948.3 1302.6

Table 2 Evaluation of results of analyses in the spherical roller bearing outer ring

Design nr. Title lef (mm) bef (mm) po (MPa) σmax (MPa) Reference Profile 21.8 0.8 1758.4 1426 bearing Cylindrical surface at the point 1 22.13 0.99 3658.6 3019 of contact 2 The change of the main radius 22.14 1.21 2048.7 1319.9 3 The combination of 2 radii 22.14 1.18 1719.7 1315.3 4 Logarithmic curve 22 0.64 1576.9 1302.6 rolling element was based on comparison of the contact bearing raceways of bearing rings was obtained in all the pressures on bearing raceways of the inner and outer designs of a new geometry of the rolling element [7-8]. bearing rings. As shown in Figures 2 and 3, the lowest contact The comparison of curves of the rolling elements pressure acts between the two bearing rings and the contact pressures depends on the length of the contact rolling element with new geometry 4. At the same time, surface lef. Figure 2 (inner ring) and Figure 3 (outer ring) the contact pressure that acts between the bearing rings show the curves’ shape. A decrease of contact pressure on and the rolling element does not produce maximum strain

COMMUNICATIONS 2/2020 VOLUME 22 OPTIMIZATION PROCEDURE OF ROLLER ELEMENTS GEOMETRY WITH REGARD TO DURABILITY... 71

Figure 4 Comparison of calculated lifetimes of analysed geometries

values that negatively affect the bearing durability. The Comparison of the calculated lifetimes of analysed new geometry of the spherical roller bearing formed by the geometries is shown in Figure 4 [10]. logarithmic curve is the most appropriate for optimization of the spherical roller bearing regarding its durability and lifespan [8]. 5 Conclusions Evaluation and selection of the most appropriate design of the new geometry are shown in Table 1 and Table Spherical roller bearings can be optimized by 2, respectively. modification of the geometry of the rolling element, i.e., the For the better evaluation of analysed geometries spherical roller. The most appropriate geometry seems to be durability of individual analysed geometries was calculated the one formed by the logarithmic curve after a comparison according to the Lundberg-Palmgren theory: of lifetimes of the bearing with the new geometry of

e c the rolling element (Figure 4). The logarithmic curve is 1 NV))x0 ln . A ) , (1) described by equations, while in this case a parameter s zh 0 of a loss of the logarithmic curve profile, i.e., a modified x0 . 0.256 ) po , (2) surface of the spherical roller. The optimal value of the parameter is 0.00035 mm which is similar as in the case of zb0 . 02. 52) , (3) the rolling bearing with the logarithmic profile. The new geometry of the rolling bearing composed by where: S is the probability of survival, the logarithmic curve increases the total carrying capacity N is the number of load cycles, and thus bearing durability by more than 25 %. This new V is the stressed volume, geometry does not form the strain peaks that negatively e, c, h, A are material constants defined by experiments, affect the total bearing durability. p0 is the pressure present at the contact point, 2b is the minor axis of the ellipse [9]. Calculation of a total lifetime of the bearing was based Acknowledgment on partial lifetimes of bearing rings. As far as a logical comparison is concerned, 100 % is assigned to the reference This work was supported by the Grant system of the geometry. University of Zilina and by Ministry of Education, Science, In a new design 1 (Cylindrical surface at the point of Research and Sport under the contract No. 1/0595/18 - contact), the calculated pressures are well above those of Optimising the internal geometry of roller bearings with line the reference bearing. Therefore, the analysis of the new contact in order to increase their durability and reduce their design 1 is excluded from further comparison. structural weight.

VOLUME 22 COMMUNICATIONS 2/2020 72 STEININGER et al.

References

[1] LUNDBERG, G., PALMGREN, A. Dynamic capacity of rolling bearings. Acta Polytechnica Scandinavica, Mechanical Engineering Series. 1952, 2(4), p. 96. ISSN 0001-687X. [2] STASTNIAK, P., MORAVCIK, M, BARAN, P., SMETANKA, L. Computer aided structural analysis of newly developed railway bogie fram. In: 21st Polish-Slovak Scientific Confeerence on Machine Modelling and Simulations 2017: proceedings. 2017. ISSN 2261-236X, p. 1-10. [3] SKYBA, R., HRCEK, S., SMETANKA, L., MAJCHRAK, M. Stiffness analysis of slewing bearings. In: 10th International Conference Machine and Industrial Design in Mechanical Engineering: proceedings. Bristol: IOP Publishing, 2018. ISSN 1757-8981, eISSN 1757-899X, ISBN 9781510868649, p.1-7. [4] BELORIT, M., HRCEK, S., SMETANKA, L. Mathematical algorithm for calculating an optimal axial preload of rolling bearings with the respect to their life. In: 10th International Conference Machine and Industrial Design in Mechanical Engineering : proceedings. Bristol: IOP Publishing, 2018. ISSN 1757-8981, eISSN 1757-899X, ISBN 9781510868649, p 1-10. [5] SKYBA, R., HRCEK, S., SMETANKA, L., MAJCHRAK, M. Strength analysis of slewing bearings. In: TRANSCOM 2019: proceedings. 1. ed. Amsterdam: Elsevier Science, 2019. ISSN 2352-1465, p. 891-897. [6] MRUZEK, M., GAJDAC, I., KUCERA, L., BARTA, D. Analysis of parameters influencing electric vehicle range. Procedia Engineering [online]. 2016, 134, p. 165-174. ISSN 1877-7058. Available from: https://doi.org/10.1016/j.proeng.2016.01.056 [7] GUO, Y., PARKER, R. Stiffness matrix calculation of rolling element bearings using a finite element/contact mechanics model. Mechanism and Machine Theory [online]. 2012, 51, p. 32-45. ISSN 0094-114X, eISSN 1873-3999. Available from: https://doi.org/10.1016/j.mechmachtheory.2011.12.006 [8] FATURIK, L., TRSKO, L., HRCEK, S., BOKUVKA, O. Comparison of structural design in high and ultra-high cycle fatigue regions. Transactions of FAMENA [online]. 2014, 38(4), p. 1-12. ISSN 1333-1124, eISSN 1849-1391. Available from: https://hrcak.srce.hr/135872 [9] STASTNIAK, P., MORAVCIK, M, SMETANKA, L. Investigation of strength conditions of the new wagon prototype type Zans. In: 23th Polish-Slovak Scientific Confeerence on Machine Modelling and Simulations: proceedings. 2019. ISSN 2261-236X, p. 1-11. [10] HARRIS, A. Advanced concepts of bearing technology. USA: CRC Press Taylor and Francis Group, 2006. ISBN 10 0-8493-7182-1.

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.73-78 73

Branislav Dobrucky - Slavomir Kascak - Michal Prazenica - Roman Konarik BASIC COMPARISON AND EVALUATION OF FUNCTIONALITY AC-AC MATRIX CONVERTER CONCEPTS FOR HEV VEHICLE - PART II

The paper deals with a modeling and simulation of the direct AC-AC propulsion system and compares two matrix converter concepts with five-phase traction induction motors (IM) for the hybrid electric vehicle (HEV). The simulation results of [3x5] matrix converter and 4Q-converter are done using Matlab-Simulink environment. Part I deals with a theoretical study of converter concepts for hybrid electric vehicle, since the configurations of [3x5]+[0x5] matrix converters with five-phase motor(s) have not been analyzed so far. Based on simulation results the comparison and evaluation of the property and quality of the quantities of different type of the matrix powertrain are discussed in Part II. Keywords: AC/AC powertrain, 3x5 matrix converter, 0x5 matrix converter, five-phase induction motor, electric drive, 4QC converter, modeling and simulation, HEV vehicle

The article is a continuation of Part I [1]. AC/AC powertrain with internal-combustion engine and synchronous generator together with a five phase induction machine and matrix converter are depicted in Figure 2. 1 Introduction The simulation results of the proposed powertrain are shown in Figures 2-4. Autonomous drive mode of HEV The paper [2] compares a [3x3] matrix converter (MxC) powered by ICE/SG is shown in Figure 3a-e. It can be seen and 3-phase VSI converter with an active front end for a as waveforms of input/output quantities of [3x5] MxC as 7.5 kW induction motor drive. It has been shown that the phase motor current/voltage and generator current/voltage, matrix converter’s semiconductor losses are smaller at full respectively. The entire control scheme in Matlab/Simulink load operation for the same silicon area in both converters. environment is shown in Figure 2 - where matrix converter A one-third reduction of the device current rating of the is presented by block 3x5 MxC. MC is possible, resulting in comparable thermal device There are simulation results of the start-up of AC/ stress. The overall passive component count and rating are AC powertrain in Figure 3a and 3c. After start-up of ICE/ only slightly better for the MxC but the absence of bulky SG during 0-2 sec with one no-load IM, the IM is loaded by smoothing capacitor is evident what emphases also work torque equal 20 Nm. Steady-state results of voltages and [3]. currents of SG and IM are given in Figure 3b and Figure A novel enhanced AC/AC series/parallel HEV 3d-e. They show a good quality of the quantity waveforms, powertrain has been introduced in Part I of [1], Figure 1. mainly of phase-current of IM. Any special control of IM has An indirect space-vector modulated three-phase not been used. AC-DC matrix converter for hybrid electric vehicles in [4], Autonomous drive mode of HEV powered by AB and with improved efficiency in [5]. The configurations of accu-battery using 4QC and [3x5] MxC is shown in Figure [3x5]+[0x5] matrix converters with five-phase motor(s) are 4. Wave-forms of input/output quantities of 4QC and [3x5] not analyzed in available literature so far. MxC can be seen there. There are results of the start-up of AC/AC powertrain powered by accu-battery and 4QC converter in Figure 4a-c 2 Comparing simulation results of both concepts and Figure 4f-g. The total course of the start-up is similar of MxC propulsion powertrain to the previous one (in Figure 3c) but DC current is taken from accu-battery AB and input currents of [3x5] MxC are Modeling approach of PMSG generator was taken different due to 4QC operation. During start-up, the IM is from [6] and [7]. Modeling of matrix converter with no-loaded. The steady-state voltage of 4QC, phase-voltage, indirect control and supposing multi-phase commutation and current IM are given in Figure 4e and Figure 4h. Phase- was adapted for five-phase from [8-11] without special current of IM (Figure 4h) is similar to that of IM powered by IM control. Model of five-phase IM can be found and AB without 4QC (Figure 5e - the next). Any special control taken from [12]. The connection of the control scheme of of IM has not been used.

Branislav Dobrucky*, Slavomir Kascak, Michal Prazenica, Roman Konarik Department of Mechatronics and Electronics, Faculty of Electrical Engineering and Information Technology, University of Zilina, Slovakia *E-mail of corresponding author: [email protected]

3x5 ICE SG M1 MxC

+ AB M2 - VCS

DIR POW Figure 1 Novel enhanced AC/AC series/parallel HEV with one MxC converter and two traction motors M1, M2 with independent control

Conn1

Discrete Conn2 1e-06 s. Conn3 powergui1 Conn4 Conn4 Conn1 TL m Conn5 Ia i Conn5 + w Conn6 - A+ dc A- s Conn1 Conn4 +

+ CM1 A Conn7 - m CVS F+ F- B Conn2 Conn5 Conn2 Conn6 Conn8 CS1

s - s C floating protection DC Machine Conn3 Conn6 isa + Ia Permanent Magnet Vai Va vas - Filter Conn7 isb Ua Synchronous Machine Vbi Vb vbs CS2

314 w* + + isc s CM Conn3 Vci Vc vcs + CVS1 - i isd - Constant1 Control Conn8 Vdi Vd vds s - ise CS3 Vei Ve ves If [3x5]MxC s omegar + Voltage measurement T_load If - Te Uf CS4 INDUCTION wf 185 MOTOR TL1 Field weakening wf

Figure 2 The control scheme of AC/AC powertrain with ICE/SG unit, [3x5] matrix converter and 5-phase IM in Matlab/Simulink environment

60 80 i i a s y n c h s a

i i 40 b s y n c h 60 s b

i i c s y n c h s c 40 i s d 20 i s e 20

0 0 I [A] I [A]

-20 -20

-40 -40 -60

-60 -80 0.5 1 1.5 2 2.5 0.5 1 1.5 2 2.5 time [s] time [s] Figure 3a Start-up of SG - phase-currents Figure 3c Start-up of IM - phase-currents

150 30 250

V u a s y n c h s a 200 i u 100 a s y n c h 20 s b 150 u s c u 100 s d 50 10 u s e 50

0 0 0 I [A] V [V] U [V] -50 -50 -10 -100

-150 -100 -20 -200

-150 -30 -250 2.1 2.11 2.12 2.13 2.14 2.15 2.5 2.505 2.51 2.515 2.52 2.525 2.53 time [s] time [s] Figure 3b The SG - phase-voltage/current in details Figure 3d The IM phase-voltages of IM

COMMUNICATIONS 2/2020 VOLUME 22 BASIC COMPARISON AND EVALUATION OF FUNCTIONALITY AC-AC MATRIX CONVERTER CONCEPTS... 75

250 20 v 200 s a i 15 s a 150 10 100

5 50

0 0 I [A] V [V] -50 -5

-100 -10 -150

-15 -200

-250 -20 2.5 2.505 2.51 2.515 2.52 2.525 2.53 time [s] Figure 3e The IM phase-current and voltage in steady-state in detail

200 30 I I A B a 4 Q C I 150 20 b 4 Q C I c 4 Q C

100 10

50 0 I [A] I [A]

0 -10

-50 -20

-100 -30 0 0.2 0.4 0.6 0.8 1 0.8 0.805 0.81 0.815 0.82 time [s] time [s] Figure 4a The current taken from accu-battery Figure 4d The 4QC phase-currents in detail

350 200

U V A B a 4 Q C 150 V b 4 Q C V c 4 Q C 100

300 0 U [V] U [V] -50

-100

-150

250 -200 0 0.2 0.4 0.6 0.8 1 0.8 0.805 0.81 0.815 0.82 time [s] time [s] Figure 4b The accu-battery voltage Figure 4e The 4QC phase-voltages in detail

150 i I s a a 4 Q C 100 i I s b 100 b 4 Q C i I s c c 4 Q C i 50 s d i 50 s e Best Position 0 0 I [A] I [A]

-50 -50

-100 -100

-150 0 0.2 0.4 0.6 0.8 1 0 0.01 0.02 0.03 0.04 0.05 time [s] time [s] Figure 4c Start-up 4QC currents of traction mode Figure 4f Start-up currents of IM

VOLUME 22 COMMUNICATIONS 2/2020 76 DOBRUCKY et al.

300 300 20

u u s a s a u i 15 200 s b 200 s a u s c u 10 s d 100 u 100 s e 5 Position

0 0 0 I [A] U [V] V [V]

-5 -100 -100

-10

-200 -200 -15

-300 -300 -20 0 0.01 0.02 0.03 0.04 0.05 0.9 0.905 0.91 0.915 0.92 0.925 0.93 time [s] time [s] Figure 4g Start-up voltages of the IM motor Figure 4h The phase voltage and current in detail

20 300

I u A B s a 15 u 200 s b u s c 10 u s d 100 u 5 s e

0 0 I [A] U [V] -5 -100

-10

-200 -15

-20 -300 0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.01 0.02 0.03 0.04 0.05 time [s] time [s] Figure 5a The current taken from AB accu-battery Figure 5d The stator voltages of the IM motor

330 300 20 U A B u s a i 15 200 s a

10 325 100 5

0 0 U [V] I [A] V [V]

320 -5 -100

-10

-200 -15 315 0 0.5 1 1.5 2 2.5 3 3.5 4 -300 -20 time [s] 0.9 0.905 0.91 0.915 0.92 0.925 0.93 time [s] Figure 5b The accu-battery voltage Figure 5e The phase voltage and current of the IM motor in details

i s a 40 100 i s b T e i s c 30 i 50 s d i s e 20

0 I [A] 10 T [Nm] -50 0

-100 -10

0 0.01 0.02 0.03 0.04 0.05 -20 time [s] 0 0.5 1 1.5 2 2.5 3 3.5 4 Figure 5c The stator currents in details time [s] Figure 5f Time course of motor torque

COMMUNICATIONS 2/2020 VOLUME 22 BASIC COMPARISON AND EVALUATION OF FUNCTIONALITY AC-AC MATRIX CONVERTER CONCEPTS... 77

250 IM

200

150 [rad/s] 100

0 0 0.5 1 1.5 2 2.5 3 3.5 4 time [s] Figure 5g Time course of angular speed

Autonomous drive mode of HEV powered by AB • battery autonomous traction and a braking regime accu-battery using [0x5] MxC is shown in Figure 5a-g. where 4QC should be completed by [3x5] MxC since The waveforms of input/output quantities of [3x5] MxC auxiliary [0x5] MxC should not be, operated as [0x5] MxC converter can be seen there. • also, charging mode from ICE/SG unit with 4QC The start-up of IM supplied by accu-battery AB is needs either control by ICE engine or further DC/DC shown in Figure 5a, Figure 5c and Figure 5e. After start-up converter since auxiliary [0x5] MxC is not needed . of ICE/SG, in time of 1, 2 and 3 sec, the IM has been loaded Total comparison and evaluation can be done until with torque 5, 10 and -10 Nm, respectively. Steady-state of when all operation modes of both concepts (with 4QC and phase voltages and currents are presented in Figure 5d and with [0x5] MxC) will be simulated and their result known. Figure 5e. Comparison to previous result mode (Figure Both of those concepts make possible to also combine 4a-c) shows that start-up of IM is faster, and shape of phase- (parallel) modes of the HEV powertrain. voltage of IM slightly different one but the phase-current is Proposed paper provides detailed study of the slighly better. Torque and angular speed circumstances are functionality of ac-ac matrix converter serving within the depicted in Figure 5f-g. modified power train of the HEV vehicle concept. Due to Comparison of autonomous modes in Figure 4 and high degree of complexity, the presented approach shows Figure 5 shows that the concept of AC/AC powertrain with mostly the simulation analysis and comparison between 4QC features: front-end VSI propulsion system and ac-ac matrix converter worse quality of quantities waveforms as accu-battery system, and comparison of two variants MxC: with- and current and output phase current and voltage of the IM without 4QC battery converter. motor Experimental verification will be realized within future worse energetic efficiency because of adding of 4QC research tasks based on the received results of expected into the power chain. behavior [13-14]. These results will be consequently A similar situation is regarding to charging of accu- published as a separate scientific paper. battery by concept with 4QC, and starting-up of ICE/ SG module using the 4QC unit or [0x3] and [5x3] MxCs, respectively. Comparison of autonomous modes in Figure Acknowledge 4 to previous Figure 2 and Figure 3 it should be noted that 4QC converter and input - 3-phase - LC filter in front of This work was supported by projects: APVV-15-0571 matrix converter are not necessary, therefore, the efficiency Research of the optimum energy flow control in the electric can be better. vehicle system, ITMS project No. 26210120021 co-funded from EU sources and European Regional Development Fund, and project KEGA No. 027ZU-4/2018. 3 Conclusion

Regarding comparison of [3x5] matrix converter system with auxiliary 4QC and auxiliary [0x5] MxC the second one is clearly better in the:

VOLUME 22 COMMUNICATIONS 2/2020 78 DOBRUCKY et al.

References

[1] DOBRUCKY, B., KASCAK, S., PRAZENICA, M., KONARIK, R. basic comparison and evaluation of functionality AC-AC matrix converter concepts for HEV vehicle - part I. Communications - Scientific Letters of the University of Zilina [online]. 2020, 22(1), p. 77-83. Available from: http://komunikacie.uniza.sk/index.php/communications/article/ view/1536 [2] BERNET, S., PONNALURI, S., TEICHMANN, R. Design and loss comparison of matrix converters, and voltage-source converters for modern AC drives. IEEE Transactions on Industrial Electronics [online]. 2002, 49(2), p. 304-314. ISSN 0278-0046, eISSN 1557-9948. Available from: https://doi.org/10.1109/41.993263 [3] WHEELER, P. W., CLARE, J.C., EMPRINGHAM, L. A Vector controlled MCT matrix converter induction motor drive with minimized commutation times and enhanced waveform quality. In: IEEE Industry Applications Conference : proceedings [online]. 2002. ISBN 0-7803-7420-7, ISSN 0197-2618, p. 466-472. Available from: https://doi.org/ 10.1109/IAS.2002.1044127 [4] ZHUANG, X. An indirect space-vector modulated three-phase AC-DC matrix converter for hybrid electric vehicles. Energy Procedia [online]. 2015, 75(4), p. 1968-1974. eISSN 1876-6102. Available from: https://doi.org/10.1016/j.egypro.2015.07.242 [5] DOBRUCKY, B., KASCAK, S., PRAZENICA, M., JARABICOVA, M. Improving efficiency of hybrid electric vehicle using matrix converters. Elektronika ir Elektrotechnika [online]. 2019, 25(4), p. 29-35. ISSN 1392-1215, eISSN 2029-5731 Available from: https://doi.org/10.5755/j01.eie.25.4.23967 [6] Sladecek, V., J. Drapela, P. Palacky, J. Machacek, D. Slivka, P. Hudecek . In: 11th International Scientific Conference on Electric Power Engineering, Brno, Czech Repuplic : proceedings. 2010. ISBN 978-80-214-4094-4, p. 249-252. [7] PowerLib Library. Matlab/Simulink. Matworks, 2017. [8] LIVINT, G. H., HORGA, V, RATOI, M., ALBU, M. Control of hybrid electrical vehicles, electric vehicles - modelling and simulations. InTech, 2011. ISBN 978-953-307-477-1. [9] DOBRUCKY, B., KASCAK, S., PRAZENICA, M., DRGONA, P., PAVLASEK, P. AC/AC powertrain control under different HEV supply network. 2018 ELEKTRO : proceedings [online]. IEEE, 2018. p. 2811-2823. Available from: https://doi.org/10.1109/ELEKTRO.2018.8398272 [10] JUSSILA, M., TUUSA, H. Comparison of direct and indirect matrix converters in induction motor drive. 32nd Annual Conference of the IEEE Industrial Electronics Society IECON’06 „ proceedings [online]. IEEE, 2006. eISBN 978-1-5090- 9155-3, ISSN 1553-572X, p. 1621-1626. Available from: https://doi.org/10.1109/IECON.2006.347423 [11] DOBRUCKY, B., KONARIK, R., STEFANEC, P., CHERNOYAROV, O. V. Modelling of transient phenomena of complex electrical circuits under periodic non-harmonic converter supply. Electrical Engineering-Archiv fur Elektrotechnik [online]. 2017, 99(4), p. 1429-1438. ISSN 0948-7921, eISSN 1432-0487. Available from: https://doi.org/10.1007/s00202-017-0658-4 [12] ZASKALICKY, P. Mathematical model of a five-phase pentacle connected IM supplied by a VSI inverter with PWM output voltage control. In: IEEE 27th International Scientific Conference Electronics ET 2018 : proceedings [online]. 2018. ISBN 978-1-5386-6692-0. Available from: https://doi.org/10.1109/et.2018.8549582 [13] KUCERA, L., GAJDOSIK, T., GAJDAC. I., MRUZEK, M., TOMASIKOVA, M. Simulation of real driving cycles of electric cars in laboratory conditions. Communications - Scientific Letters of the University of Zilina [online]. 2017, 19(2), p. 42-47. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/ article/view/210 [14] HUSAIN, I. Electric and hybrid vehicles - design fundamentals. 2. ed. CRC Press, 2011. ISBN 978143981175.

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.79-87 79

Libor Hadacek - Lenka Sivakova - Radovan Sousek - Mikael Zeegers ASSESSMENT OF SECURITY RISKS IN RAILWAY TRANSPORT USING THE FUZZY LOGICAL DEDUCTION METHOD

The aim of the paper is to inform about the possibilities of using a fuzzy logical deduction in security practice. The fuzzy logic deduction allows to record the management experience in IF - THEN rules and does not require a precise description of the parameters of the controlled function. This property is an important asset for risk assessment in an incompletely defined environment. The application of the method is demonstrated in the security risk assessment of the physical protection of the national railway with a focus on the corridor railway lines and with regard to the future construction of high-speed railway lines in the Czech Republic. At present, it is a generally accepted fact that securing basic transport functions is a prerequisite for successful crisis management. These functions can be specified as road and rail negotiability. Keywords: fuzzy logic deduction, security, safety, rail, crisis management

1 Introduction or 0 that corresponds to states such as on and off. Fuzzy logic expands the two-logic logic to multi-valued logic. An Security management activities in an organization example of the difference between two-state logic and fuzzy involve a great amount of decision-making on strategy and logic is shown in Figure 1. tactics of risk management [1-3]. Security managers have The fuzzy logical deduction is based on the basic to make decisions despite uncertainties, inaccuracies, or rule of human logical thinking. Deduction, i.e. drawing incompleteness of input data [4-5]. They use quantitative conclusions, is implemented through formulas. The most tools to make decisions that require a certain level of input important rule used is the modus ponens rule. Formally, it data precision. Furthermore, they can also apply semi- is possible to write it as follows: quantitative analytical methods based on a point scale. In AA, & B the case of more independent variables, their point values B . (1) are usually obtained using binary operations. This results in the value of the output dependent variable being the same In this rule, there are A, B formulas. The rule modus for a different combination of values of the input variables. ponens says that if we know the fact labeled by formula A possible method of making decisions with uncertainty A and we know that the fact B is based on fact A, then we or incompleteness of input data is the application of can assume that the fact B is valid. In classical logic, true fuzzy logic. Fuzzy logic isused in controlling processes of and false formulas are examined in models, while fuzzy industrial products. For example, their principles control logic examines formulas whose truth value in models the image processing processes or automatic washing is different [7]. The principle of fuzzy logic deduction machines programs. is illustrated by the general fuzzy controller shown in To control processes in an incompletely defined Figure 2. The general fuzzy controller consists of blocks of environment, it is appropriate to use a fuzzy logical fuzzification, knowledge base, inferential mechanism, and deduction. Fuzzy logical deduction (FLD) is a part of fuzzy defuzzification. logic in a broader sense. FLD was described by prof. V. In the fuzzification process, the input independent Novak in [6] in 1995 with the aim of creating an exact formal variable is assigned a language expression. It is a variable theory. whose values can be words or some natural language expressions [7]. An example of the names of selected language expressions is given in Table 1. 2 Model of risk assessment using the fuzzy logical The extensions of evaluation predictions are deduction method constructed identically in all contexts. Typical elements have a degree of competence in the given extension of the In common life, an event can be described by the two- evaluation prediction equal to 0.9 or 1. The example of the state logic. Individual values can be assigned a value of 1

Libor Hadacek1,*, Lenka Sivakova2, Radovan Sousek1, Mikael Zeegers3 1Faculty of Transport Engineering, University of Pardubice, Czech Republic 2Faculty of Security Engineering, University of Zilina, Slovak Republic 3Security Risk Watch, Voorthuizen, Netherlands *E-mail of corresponding author: [email protected]

Figure 1 Difference of two stage logic and fuzzy logic

Figure 2 General scheme of fuzzy controller [7]

Figure 3 Extension of evaluation predictions [7]

Table 1 Names of selected language expressions [7] Table 2 Example of IF-THEN rules

Number Languge expression Abbreviation Number X1 is Ai & X2 is Bi => Y1 is Ci 1 Small Sm 1 sm Ze | ze 2 Roughly Small Ro sm 2 sm Sm | sm 3 Midle Me 3 sm ro sm | ro sm 4 Roughly Big velky Ro Bi 5 Big Bi course of selected extensions of the evaluation predictions the input variables based on the results of the controlled for the context 04,,0 100 is given in Figure 3 [7]. process observation. An example of creating rules is given A set of rules is stored in the knowledge base. An in Table 2. approximate knowledge of the regulatory strategy is In the inference mechanism, logical deduction is sufficient to create the rules. The strategy is described using applied in the decision on the input variables according to the IF-THEN fuzzy set of rules [7]: the respective rule. The conversion of a language expression into a real

RIn = FXis ATHENYis B , (2) number occurs in the defuzzification process. Defuzzification is an operation that assigns an element from its carrier to where: the fuzzy set.

Rn is the rule and n is the rule number, X is the assessed object, DEFA ! Supp A . (3) ^^hh A is object property, Y is a dependent variable, For defuzzification, the method of DEE (Defuzzification B is object property. of Evaluative Expressions) is applied in combination with In order to develop a controlling strategy, the the fuzzy logical deduction. The reason for using it is expressions of a natural language, whose meaning is a smoother course of the resulting function. The method is generally known, are used. The individual rules are made described in detail in [8]. up of vague statements characterizing the properties of

COMMUNICATIONS 2/2020 VOLUME 22 ASSESSMENT OF SECURITY RISKS IN RAILWAY TRANSPORT USING THE FUZZY LOGICAL DEDUCTION... 81

Figure 4 The value of the risk is a function of an event probability and an asset loss consequence

Figure 5 The model of fuzzy controllers system for determining of the risk of an asset loss

Determining the value of the risk by the FLD method Fuzzy controllers can be connected together. This is based on the same principles as the traditional method feature allows you to solve a complex task when only one of determining the risk value, as described in the technical set of rules is not sufficient. Assessing security risks is standard [9]. The risk values presented in Figure 4 are based a complex task, the result of which supports the decision- on Table 5. The method of determining the values changes making process of security management on the use of risk the established linear perception of the course of risk management strategies. Security risks are, by their nature, values. The created lines of risk are similar to those of the intentional, thoughtful and purposeful acts of a person output characteristics of the transistor. who implements a scenario of an attack [10]. Due to The proposed procedure eliminates the lack of the uniqueness of the location of the protected object, a traditional risk-setting method. For example, with a number of uncertainties and incompleteness of the input a probability value of 0.1 and the consequence value of 100, data can be considered. A model of the fuzzy controller the resulting value of risk is 10. If the probability is 1 and system for determining the value of the risk of an asset loss the magnitude of consequence is 10, then the risk value is is shown in Figure 5. again 10. When applying the FLD method to determine the The asset is related to the internal and external risk values, in the first case the risk value is 23.1 and in the context. Contexts include the number of casualties. Other second case, the result is 14.7, as can be seen in Table 5. contexts include damage to property and the environment. The time context is the period of recovery of the asset

VOLUME 22 COMMUNICATIONS 2/2020 82 HADACEK et al.

to its original state after the effects of the scenario. The e) Description of the asset - where other asset properties level, scope, and effectiveness of asset risk management are specified. measures are influenced by the scenario. The relationship between measures and scenarios is applied in an expert judgement on asset vulnerability. A measure of the asset‘s 3.2 Scenario vulnerability is the probability of its loss. The consequence of the loss of an asset is a measure of the asset‘s context. The scenario is a set of conditions and/or events that can The risk of loss of an asset depends on the probability of cause a security incident. A deliberate anthropogenic threat loss of the asset and the consequence of its loss. is an individual or group of individuals with motivation and For the purposes of the risk assessment procedures ability to act deliberately to cause loss or damage to the outlined in this text, we applied definitions of the event, asset. The method of the threat manifestation is described occurrence, possibility or probability of occurrence, risk, by the scenario. The list of threats is specified in the threat and risk levels as set out in the technical standard [11]. catalog. When creating the threat catalog, legal provisions Vulnerability means the property of any material and the experience of experts from previous events are object, technical means or social entity to lose the ability to used. For each scenario, the following characteristics are fulfill its natural or established function due to external or given: internal threats of different nature and intensity. [12] a) Threat, It can be concluded that the vulnerability characterizes b) Direction of an attack - it is important for the choice of the ability of the physical protection system to prevent preventive measures, the loss, damage or destruction of the protected object. c) Time Attributes (seasons, time, opening hours), It qualitatively or quantitatively expresses the degree of d) Method attributed (tools, weapons, vehicles), probability that something will happen, i.e. probability of e) Description - other data supporting the expert judgment loss of a protected subject. It follows that the probability of on the vulnerability of the asset. asset loss is a function of vulnerability.

3.3 Installed measures 3 Assessment of the risks of physical protection of high-speed rail For the purpose of physical protection, it is important to know the security features used to deter, slow, detect, The software application „Analysis of Specific Risks and interrupt the attacker‘s progress. In particular, these of Physical Endangerment of the High-Speed Railway include the construction of windows, doors, roof skylights, Infrastructure“ was developed for the solution of the ventilation ducts and other openings of the building shell security risks of high-speed railway lines using the fuzzy that can facilitate unauthorized entry [16]. For buildings logical deduction method. For the purposes of this paper, with a perimeter, knowledge of fences, gates, barriers etc. the issue of high-speed lines has been reduced to the level is important. Details of current measures are listed in the of corridor tracks where the speed is 160 km/h. following fields: a) Asset - the name of the asset, b) Safety / Security region - the area of safety or security, 3.1 Asset e.g. physical protection, c) Type - general designation of the measure, eg. fencing, According to the technical standard [13], the asset is d) The title of the measure - closer measure specification, anything that has value for the organization. The definition e) Level - the quality of the measure, e.g. burglary can be extended to the importance of the asset for its resistance, owner. The types of railway infrastructure assets can be f) Place of use - name of the security barrier, e.g. considered, for example, a bridge, a tunnel [14], a station perimeter, head, a grade separation structure or a track section [15]. g) Purpose - which part of the security system is protected, Assets are entered into the database manually or from pre- h) Date of commissioning - significant for systems prepared xls files. degrading with time, For the purpose of assessing the risks of physical i) Description - a more detailed description of the protection of the high-speed rail infrastructure, the measure, the item is not indexed, infrastructure asset is characterized by the following fields: j) Photo - this item allows you to attach a digital photo to a) Type - general designation of the group of assets, the description. b) Asset name - a closer unique identification of the asset, c) Asset location - place, premises, space (regional headquarters, track section, ID), 3.4 Asset context d) GPS coordinates - the location of the asset in map coordinates, The process of determining the asset context, based on the technical standard [11], consists of subprocesses:

COMMUNICATIONS 2/2020 VOLUME 22 ASSESSMENT OF SECURITY RISKS IN RAILWAY TRANSPORT USING THE FUZZY LOGICAL DEDUCTION... 83

Table 3 Formalized vulnerability sentences and language expressions Number Semantics of Vulnerability Language expression Abbreviation 1 Not assessed Zero Ze 2 Protection multiply exceeds the requirements Small Sm 3 Protection exceeds the requirements Roughly small Ro Sm 4 Protection meets the requirements Very Roughly small VR Sm

Table 4 Groups of risk values Number Groupe title Extent 1 Low risk 0.0-30.0 2 Moderate risk 30.1-36.9 3 Very high risk 37.0-60.0 4 Extreme risk 60.1-100.0

Table 5 Risk values determined from the characteristic values of the input variables [18] 1.0 0.0 14.7 21.3 30.2 39.9 50.0 60.0 69.9 79.8 88.2 95.6 0.9 0.0 14.7 21.3 30.2 39.9 50.0 60.0 69.9 79.8 86.8 88.3 0.8 0.0 14.7 21.3 30.2 39.9 50.0 60.0 69.5 71.1 76.5 78.0 0.7 0.0 14.7 21.3 30.2 39.8 49.7 52.8 56.3 57.1 61.9 65.2 0.6 0.0 14.7 21.3 30.2 39.8 47.2 48.6 50.6 50.9 54.0 56.3 Asset loss 0.5 0.0 14.7 21.3 30.2 39.4 43.7 44.1 45.0 45.0 46.6 47.6 probability 0.4 0.0 14.7 21.3 30.2 39.0 41.3 41.3 41.3 41.3 41.3 41.3 0.3 0.0 11.7 20.2 29.8 35.2 36.9 36.9 36.9 36.9 36.9 36.9 0.2 0.0 11.7 19.8 27.9 27.9 31.1 31.1 31.1 31.1 31.1 31.1 0.1 0.0 11.7 18.3 19.8 19.8 23.1 23.1 23.1 23.1 23.1 23.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 10 20 30 40 50 60 70 80 90 100 Asset losss consequence

determining contexts and determining risk criteria. The vulnerability analysis results, including associated language context is defined as the goal the organization seeks to expressions, is given in Table 3. achieve in its internal and external environments. The In the risk criteria identification subprocess, the external and internal contexts justify the reason why organization, besides other aspects, determines the levels preventive measures are planned and applied for the asset. of risk. According to the technical standard [17], the For the purpose of assessing the risks of physical risk levels can be divided into three groups (low, large, protection, the targets are set as the number of casualties, extreme), four groups (low, medium, very large, extreme) the economic impact on property and the environment and or five groups (low, medium, large, very large, extreme). For the time necessary to restore the asset‘s function. External the purpose of assessing the risks of physical protection, it target values are as the same as internal target values, exept is appropriate to use the four groups of risk levels, as listed restoration. Restoration in external means the restoration in Table 4. of damage in external area or the reputation of the asset A four-level risk classification allows security owner. The organization considers the importance of goals management to select an appropriate security strategy for in both environments to be equivalent. risk management. In the created SW application, three experts separately assess the level of achieving the context objectives of the context. Answers are formulated in numerical ranges or 3.5 Security risk assessment a formalized sentence. By selecting a predefined answer, it is possible to answer the question „what maximum value The resulting risk value is the output of the fuzzy of the given context can the attack scenario achieve?“ controller - the value of the risk. The rules for the An example of formalized sentences for recording the determination of risk values using FLD were adopted

VOLUME 22 COMMUNICATIONS 2/2020 84 HADACEK et al.

Figure 6 Results of risk analysis

Table 6 Summary of links of terms for risk solution Risk treatment Example Risk treatment strategy restriction, mitigation, transfer, preservation, avoidance Safety / Security region Physical Protection, Fire Protection, Information Security Measures group routine security precautions, physical security etc. Typ of measure fence, gate, door, PIR detector, detection camera, site security Safety / Security element door RC 3, switchboard 3 Reason of measure mechanical barriers, detection systems, camera systems from [18], where was for the calculations used the LFLC1 zone. The y-axis shows the numbers of the scenarios under program. The risk values determined by the FLD method consideration. are shown in Table 5. Splitting the chart into two parts and coloring the The input variable „Consequence of the asset loss“ and groups of risk values makes the significance of the risks the input variable „Probability of asset loss“ are defined by evaluated more transparent. The risk values groups are the characteristic values corresponding to the range of the color-coded in accordance with the technical standard [17]. minimum and maximum risk values. 3.7 Risk treatment

3.6 Risk evaluation In the process of risk treatment, measures are implemented with regard to risk treatment strategies The last step of the risk assessment, according to [11], selected in the Risk Assessment process. Risk treatment is is the risk evaluation. An example of a possible graphical carried out through one of the strategies [19]: representation of the risk values is shown in Figure 6. a) Restrictions - the use of preventive measures to The x-axis shows the risk values. Risk values with minimize the probability of loss of the asset and the a „-“ sign are not mathematical. The use of a minus sign consequences of loss of the asset, helped to separate the risks of vector scenarios from the b) Mitigation - limiting all negative consequences of an external environment, where the risk source is outside the event, protected zone, from other risk vector groups. Positive risk c) Transfer - sharing the cost of losses with another entity, values are in the graph represented by risk vectors from the d) Preservation - knowingly accepting the costs of losses, inside out, from the protected to the unprotected zone and e) Avoidance - the non-possession of the asset to which the vectors of internal risks, i.e. risks within the protected the risk relates. Each of the risk treatment strategies includes at least one of the security area. More than one strategy and more 1 LFLC – Language Fuzzy Logic Controler developed in the than one security area can be selected. Security areas Institute for research and Application of Fuzzy Modeling, consist of groups of measures. There are different types University of Ostrava, http://irafm.osu.cz/en/c100_software/

COMMUNICATIONS 2/2020 VOLUME 22 ASSESSMENT OF SECURITY RISKS IN RAILWAY TRANSPORT USING THE FUZZY LOGICAL DEDUCTION... 85

Figure 7 Interpretation of the risk assessment results in the map of measures in the group of measures. Security features 3.8 Interpretation of results of the security risk marked with an exact security feature specification are assessment assigned to the type of measure. Each type of measure has a specific purpose in the physical protection system. The results of security risk assessment are interpreted A summary of the concept links is given in Table 6. in the form of text, tables, charts, drawings in the map. For example, physical protection is a security area Here, all the data put in and created in the assessment from a „limitation“ strategy. The groups of measures include processes and subprocesses are applied. The interpretation technical measures, routine security precautions, and method depends on the purpose and the recipients of the physical security. The technical measures group includes data. It is necessary to consider whether this is the interim the types of measures, mechanical barriers, detection information needed for the experts or the final report to the systems, camera systems and others. Types of measures evaluator. in the group of mechanical barriers include fences, gates, A possible interpretation of the risk assessment results barriers, doors, locks and others. A door is considered as in the map is shown in Figure 7. For the assessment, three a security feature when it meets the requirements of EN tunnels were selected. The scenarios under consideration 1627 RC 3. Using the security features of the different types were divided into three vector groups. For the assessed of measures of the measure groups, a physical protection asset, the maximum risk values for each attack vector and system is created. the information on the threat level of terrorist attack by the New measures can be added to the protection system. Ministry of the Interior of the Czech Republic are clearly The current measure may be replaced or removed. To outlined. The resulting risk values and degree of threat record details of the measure proposal in the software, by a terrorist attack are demonstrative and can not be the same fields as described in Subchapter 3.3 are filled in, considered realistic. supplemented by the following fields: a) Duration - the time needed to implement the measures, b) Feature description - a more detailed description of the 4 Conclusion measure, the item is not indexed, c) Cost of measure installation - costs of installing The text summarizes the basic theoretical knowledge measures, necessary to determine the risk value by the fuzzy logical d) Cost of measure operation for one year - cost of deduction method. The example of railway transport measure operation in 1 year. infrastructure objects was used to demonstrate the The selected procedure was used to maintain procedure for the creation of the structure of fuzzy consistency between two processes in the software regulators for the determination of the risk value in the application. Appropriate costs of installing measures are field of physical protection. the costs which cover the approved organization‘s risk The described procedure is applicable by the security management scenario. management to determine the security risk values without detailed knowledge of the fuzzy logic deduction theory and software ownership. It is relatively easy to apply it in railway transport conditions. The proposed risk assessment method, using a fuzzy logical deduction, enables the risk criteria and the content

VOLUME 22 COMMUNICATIONS 2/2020 86 HADACEK et al.

of the auxiliary databases to be tailored to specific user navrh na doplnkova bezpecnostni opatreni fyzicke conditions. The maximum range of risk values and risk ochrany/ Analysis of threats of high-speed railway assessment criteria are set with respect to the asset holder‘s infrastructure by terrorist attacks and extremism, security documentation. The selected risk assessment proposal for additional security measures of physical process increases the number of potential users of the protection“, code number VH20172019016. A specific grant software application. given by the Ministry of Interior of the Czech Republic within the program of VH - Program bezpecnostniho vyzkumu pro potreby statu 2016 - 2021 (BV III/2 VZ) Acknowledgements (2016 - 2021). This article was prepared under the support of the This article was prepared under the support of project VEGA 1/0628/18 Minimizing the level of experts’ the project „Analyza ohrozeni vysokorychlostni zeleznicni estimations subjectivity in safety practice using quantitative dopravni cesty teroristickymi utoky a extrenmismem, and qualitative methods.

References

[1] ZAGORECKI, A. T., RISTVEJ, J., KLUPA, K. Analytics for protecting critical infrastructure. Communications - Scientific Letters of the University of Zilina [online]. 2015, 17(1), p. 111-115. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/402 [2] LOVECEK, T. 2008, Present and future ways of physical property protection. Communications - Scientific Letters of the University of Zilina [online]. 2008, 10(1), p. 35-39. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/1028 [3] RISTVEJ, J., ZAGORECKI, A., HOLLA, K., SIMAK, L., TITKO, M. Modelling, simulation and information systems as a tool to support decision-making process in crisis management. In: Modelling and Simulation 2013 - European Simulation and Modelling Conference ESM 2013 - EUROSIS 2013: proceedings. 2013, p. 71-76. [4] ZAGORECKI, A. T., JOHNSON, D. E. A., RISTVEJ, J. Data mining and machine learning in the context of disaster and crisis management. International Journal of Emergency Management [online]. 2013, 9(4), p. 351-365. ISSN 1471-4825. Available from: https://doi.org/10.1504/IJEM.2013.059879 [5] LOVECEK, T., VELAS, A., KAMPOVA, K., MARIS, L., MOZER, V. Cumulative probability of detecting an intruder by alarm systems. In: 47th International Carnahan Conference on Security Technology ICCST 2013: proceedings [online]. IEEE, 2013. eISBN 978-1-4799-0889-9. Available from: https://doi.org/10.1109/CCST.2013.6922037 [6] NOVAK, V. Towards formalized integrated theory of fuzzy logic. In: Fuzzy logic and its applications to engineering, information sciences, and intelligent systems. Theory and Decision Library (Series D: System Theory, Knowledge Engineering and Problem Solving). BIEN, Z., MIN, K. C. (eds.) [online]. Vol. 16. Dordrecht: Springer, 1995, p. 353-363. ISBN 978-94-010-6543-6, eISBN 978-94-009-0125-4. Available from: https://doi.org/10.1007/978-94-009-0125-4_35 [7] NOVAK,V., KNYBEL, J. Fuzzy modelovani / Fuzzy modeling (in Czech). Ostrava: Ostravska univerzita, 2005. [8] NOVAK, V., PERFILIEVA, I. On the semantics of perception - based fuzzy logic deduction. International Journal of Intelligent Systems [online]. 2004, 19(11), p. 1007-1031. eISSN 1098-111X. Available from: https://doi.org/10.1002/int.20034 [9] CSN EN 60812. Techniky analyzy bezporuchovosti systemu. Postup analyzy zpusobu a dusledku poruch (FMEA) / Analysis techniques for system reliability. Procedure for failure mode and effects analysis (FMEA) (in Czech). [10] CSN EN 14383-1. Prevence kriminality. Planovani mestske vystavby a navrhovani budov. Cast 1: Definice specifickych terminu / Prevention of crime. Urban planning and building design. Part 1: Definition of specific terms (in Czech). [11] CSN ISO 31000. Management rizik. Principy a smernice / Risk management. Guidelines (in Czech). [12] HOFRIEITER, L., BOC, J., JANGL, S., LOVECEK, T., MACH, V., SEIDL, M., SELINGER, P., VELAS, A. Ochrana objektu kriticke dopravni infrastruktury / Critical transport infrastructure protection (in Czech). 1 ed. Zilina: EDIS - Publishing House of the University of Zilina, 2013. ISBN 978-80-554-0803-3. [13] CSN ISO/IEC 27001. Informacni technologie. Bezpecnostni techniky. Systemy managementu bezpecnosti informaci. Pozadavky / Information technology. Security techniques. Information security management systems. Requirements (in Czech). [14] MOZER, V., OSVALD, A., LOVECEK, T., FANFAROVA, A., VRABLOVA, L. Fire safety in tunnels forming part of critical infrastructure. In: 47th International Carnahan Conference on Security Technology ICCST 2013: proceedings [online]. IEEE, 2013. eISBN 978-1-4799-0889-9. Available from: https://doi.org/10.1109/CCST.2013.6922042 [15] SOUSEK, R. Krizovy management a doprava/ Crisis management and transport (in Czech). Pardubice: IJP, 2010. ISBN 9788086530642.

COMMUNICATIONS 2/2020 VOLUME 22 ASSESSMENT OF SECURITY RISKS IN RAILWAY TRANSPORT USING THE FUZZY LOGICAL DEDUCTION... 87

[16] CSN EN 15602. Poskytovatele bezpecnostnich sluzeb. Terminologie / Security service providers. Terminology (in Czech). [17] ISO 22324. Societal security. Emergency management. Guidelines for colourcoded alerts. [18] HADACEK, L. Vyuziti fuzzy logiky pro studii bezpecnostnich hrozeb a rizik fyzicke ochrany / Using fuzzy logic to study security threats and physical protection risks (in Czech). Ph.D. thesis. Ostrava: VSB - Technicka univerzita Ostrava, 2015. [19] CNI Pokyn ISO/IEC 73. Management rizika. Slovnik. Smernice pro pouzivani v normach / Risk management. Vocabulary. Guidelines for use in standards (in Czech).

VOLUME 22 COMMUNICATIONS 2/2020 88 https://doi.org/10.26552/com.C.2020.2.88-96 ORIGINAL RESEARCH ARTICLE

Dusan Janostik - Viktor Nohal - Herbert Seelmann - Jaroslav Smutny THE CONTINUOUS MONITORING OF SELECTED RAILWAY STRUCTURES USING THE AUTONOMOUS DATA LOGGER

This contribution focuses on issue of the autonomous monitoring of selected dynamic parameters of the railway lines test sections under conditions in the Czech Republic. Description of the possibilities of such data collection, including the designed hardware and software, is a part of this contribution. At the same time, the contribution describes application of such a measuring system, comparing the selected dynamic parameters of the two types of switches, the first one was a switch with a standard fastening node, the other one with an elasticized fastening node. Recommendations and conclusions of railway infrastructure manufacturers and railway line managers are parts of this contribution, as well. Keywords: data logger, vibrations, monitoring, railway

1 Introduction to hold the track length in a precise geometrical position. This function is conditioned by suitable properties of gravel The railway line (or the tram line) represents a multi- grains. Their size must be optimal, the compaction must be layer system determined for the motion of railway vehicles sufficient (single grains must be mutually placed so that including the equipment necessary for ensuring the safety there may be the smallest possible space between them) and smoothness of the railway traffic. The traffic line is and they must be mutually wedged in by their sharp edges formed by the superstructure and the substructure. By [2]. the substructure is understood the railway earth bed Vibrations of the superstructure are affected by its and the structures that completely or partially substitute quality, by operation and structural conditions, climatic protective structures and equipment. Its task is to support phenomena and especially by the dynamic loading by the superstructure. The track bed forms the boundary the wheel pairs of the rail vehicles. Related to the rising between the substructure and the superstructure [1]. speed on the railway tracks, the dynamic effects acting The superstructure is formed by the railway track, on the rail structure increase, as well. These effects which carries and directs vehicles. The superstructure is negatively influence the rails and lead to the development of constructed either as a classic one with the gravel ballast failures and defects. Defects usually become evident in the or as a slab track. It should be mentioned that in the Czech disintegration of the geometric parameters of the rail, which Republic the ballast-free track represents the so-called non- consequently leads to the repeated increase of dynamic conventional structure of the superstructure. Its application effects, which has negative affect on the functionality of occurs especially in special cases like for instance is the the whole system. The result is a malfunction of the entire tram track, the underground or the track passing through system and thus also endangering the continuity and safety tunnels. The rail of the ballast-free track is mostly fixed to a of the rail traffic. It may be unambiguously said that the concrete slab or sleeper, which is fixed to the slab. It should solution of the majority of structural problems rests in the be noted that the text to follow deals with the problems right understanding of dynamic effects acting in the railway focused on the classic superstructure. track structure. The fundamental part of the superstructure is formed The basic claim put on single parts of the railway by rails, turnouts, fasteners, sleepers and the railway tract is their functional reliability together with the safety bed. Rails, together with sleepers, fasteners and the track of the railway operation. Although the superstructure fastenings, constitute the track length. Its main function has been practically developed to its perfection during is to carry the vehicle and the load, to lead the vehicle more than 150-year railways, new technical solutions can geometrically along the trajectory, and by its elasticity to still be found. This holds good especially for different absorb dynamic effects and also to distribute the load into types of turnout structures, various rail fastenings, various other layers of the superstructure and substructure. types of rail pads, sleepers, anti-vibration measures etc. The main function of the gravel ballast is to resist the It is suitable to say that each part of the rail structure vertical, cross and longitudinal forces and at the same time is exposed to considerable static and dynamic stresses.

Dusan Janostik1,*, Viktor Nohal1, Herbert Seelmann2, Jaroslav Smutny1 1Institute of Railway Structures and Constructions, Faculty of Civil Engineering, Brno University of Technology, Czech Republic 2Rail Data Services Austria GmbH & Co KG, Wien, Austria *E-mail of corresponding author: [email protected]

Figure 1 The data logger set with acceleration sensors Permanent pressure on increasing the transport speed approach. At the beginning of the Fourth Industrial and the operational load of tracks causes a magnificent Revolution, it is certainly worth considering thinking about development of new technologies in the railway traffic. the implementation of the smart railway structures, which Concerning the modernization of railway lines, this would automatically inform a main system of their status qualitative shift also refers to the Czech and Slovak railway and/or any changes in their status. tracks [3]. Decision on modernization of the lines has To achieve this state, the railways structures become an impulse to develop all the branches of the (superstructure, points, etc.) have to be provided with railway traffic both in the field of vehicles and in the field an autonomous measuring device that would perform of the infrastructure. Development and application of new continuous measurements, process the measured data and experimental procedures for evaluation of the quality and transfer it to a main evaluation system based on proposed usefulness of particular structural solutions must be in scenarios. Therefore, the authors of this contribution have accordance with this trend. developed a low-cost automatic measuring system [4]. The system allows not only the “smart” collection of basic data but includes also a set of algorithms enabling information 2 The measurement technology on status of the transport structures and their components to be gathered in the real time and stored for future use as Measurements performed on real railway structures required by an administrator or the design department and/ are very important. Field measurements take account or for monitoring of an impact on the surroundings. The of the stochastic character of effect of the train sets system has been primarily designed to monitor and evaluate and properties of the railway line construction. For that selected dynamic characteristics (vibrations, deformation). purpose, the so-called test sections have been established The monitoring system provides large data volumes, in the railway network in the Czech Republic to test new however, not all the data need to be stored. It is more useful structures of the superstructure and the substructure in to pre-process data continuously and to store only selected the long run. Demands put on test sections are usually very information. The data storage process includes algorithms strict. This applies primarily to quality of construction and/ of intelligent sorting and processing. or reconstruction, the monitoring of selected parameters, and the access and establishment of measuring points. It should be noted that a new structure to be tested is applied 3 Description of the measuring system in a track together with its reference, classic variant, i.e. construction without innovation. The entire measuring system was designed as a two- It should be mentioned that the short-term level one. The first level is a data logger, the second is an measurements have currently prevailed. This means evaluation software running on a main computer [4]. that selected parameters of the railway line have been The data logger (Figures 1 and 2) is designed as a measured for one or several days. The repeatability of those separate measurement point. It may be used throughout the measurements over a period of one year has been relatively construction of a railway superstructure with the possibility low, with the measurements being usually repeated twice a to record up to 12 analogue quantities in the form of an year. This procedure is appropriate in some cases, while for electric voltage of 0V to 5V. It is primarily intended to some structures it would be useful to apply a continuous record vibrations, noise, deformation, displacements and

VOLUME 22 COMMUNICATIONS 2/2020 90 JANOSTIK et al.

Figure 2 The in-situ view of the data logger temperature. Quantities are continuously stored on an SD The display shows bar charts of the minimum card. and maximum of all the channels before starting the The data logger is built on the microcomputer measurement. The touchscreen makes it possible to select Teensy platform with a 32-bit ARM the appropriate number of measuring channels to measure Cortex-M4 processor. It can be programmed very similarly and display averaged maximum and minimum values before to the well-known Arduino system directly via a USB measuring to calculate the sensitivity of, for example, the connector. The system is designed to capture a large accelerometer sensor. At the same time, it is possible to number of time records of any length at a sampling rate specify the limit values, whereby the logger reports to the of 100 Hz to 4000 Hz. This range of sampling frequencies main system a serious problem found in design and sends is sufficient to measure vibrations on the track structures. the data for immediate processing. During the ordinary Note that in this area the signals are evaluated up to about operation, the main system automatically selects data from 1 kHz. The Teensy microcomputer board was completed individual data loggers for further processing according to with additional components for measurement purposes. It the schedule. was mainly the part forming the measurement chain, power The electronics of the data logger is placed in a durable module, data logger synchronization module, data storage box. The Sub-D9 connectors are used to connect sensors in to SD card module and a data communication module the proposed version. The system has its battery, as well as with a main system. A substantial part of the work was the possibility of external charging/powering, for example, devoted to development of a data logger control system and by solar panels or by electricity network. The device is application software. designed to be used in real weather conditions. It can be The system is equipped with a real-time clock with the attached to the parts of the railway superstructure (under possibility to synchronize more of the same data loggers. the rail, on sleepers, etc.). Measurement and data storage can be done either on Several triaxial MEMS-type accelerometer sensors external interruptions (change of logic level using, e.g. were designed and manufactured for the data logger, two the gate trigger) or at the set signal level on the selected with a measuring range of 200 g (expected use on the measurement channel. rail), two of 16 g (assumed use on a sleeper), two with After measuring and storing the recording (time events), a switchable measuring range of 2 g/6 g (assumed use the system is set to a standby mode to wait for further outside the track skeleton, possibly in gravel). The MEMS measurement. The measuring system is equipped with a chips from the Analog Devices Company were used to touchscreen with possibility to set the basic parameters assemble the sensors supplemented with electronics to and operations – methods of starting, stopping, setting of ensure their power supply and adaptation to the data logger. measuring time, basic sampling frequency, selection of The produced data logger was tested in the laboratory measuring channels, etc. and field during the year 2019, including the comparative measurements with the Dewetron 2500 system. The

COMMUNICATIONS 2/2020 VOLUME 22 THE CONTINUOUS MONITORING OF SELECTED RAILWAY STRUCTURES USING THE AUTONOMOUS... 91

Figure 3 Demonstration of application uploaded to a mobile phone and connection between the mobile phone and a data logger measuring string was calibrated using a calibration sensor 4 The measurement description 8305 from Brüel Kjaer and a vibrating calibrator V21D from Metra Mess und Frequenztechnik. Monitored switches No. 3 and No. 4 (Figure 4) are The proposed data logger communicates with a main situated in the Trebova deviation of the railway station system in line with current and new trends using WLAN of Usti nad Orlici. Both switches are located on a high Wi-Fi radio non-cellular technology. The assumed data embankment. They have the same slenderness (1:12 – 500), transmission reaches tens of meters. both are run by most trains alongside the tip of the frog, The data logger comes with software created to allow and have the same frogs (UIC 60 superstructure, ZPT- the data transfer to the main system and their further data monoblock frog). Switch No. 4 has a standard fastening processing. Currently, the software provides the ability to using ribbed plates on concrete sleepers with Skl 24 create images with two types of charts. On the left-hand side, clamps and it is located on track No. 2. Switch No. 3 has there are the time courses of the measured quantity while a new UNO3718 fastening using ribbed plates on concrete their amplitude spectra are on the right-hand side. At the sleepers with Skl 24 clamps and it is located on track No. 1. same time, other characteristics are identified and counted, The main difference between the above fastenings is mainly e.g. global maxima and minima, RMS values, or floating in the pad under the foot of rail. Switches are placed on the RMS vector. The software database can be completed with pre-stressed concrete VPS sleepers. characteristics of locomotives and carriages. By using this The dynamic effects were measured after the frog data, it is possible to detect the passing train from the replacement on the new explosion-reinforced manganese measured records. The data read from the data logger, as frog constructions. It was done both for comparison of well as their calculated characteristics, are further stored in both constructions and for verification of measurement created structures of the SQL database. From this database, methodology based on use of autonomous data logger. the pre-processed data can be loaded into the main analysis The measurement was carried out simultaneously on both system. At the end of the data logger description, it is also switches; the measurement conditions were identical for necessary to mention the possibility to read and analyse both measured structures. the measured data using a mobile phone with the Android The selected design for measuring and investigating operating system. An application was created to read and the dynamic effects was the switch frog [5]. When a wheel analyse data over a Wi-Fi connection (Figure 3). of a train set is passing through, there is a shock and vibration in the frog. The load is distributed through the

VOLUME 22 COMMUNICATIONS 2/2020 92 JANOSTIK et al.

Figure 4 Measurement place, the view of switches No. 3 and No. 4 wheel, into the frog, into the elastic fastening, in which the frog structure on the sleeper-gravel bed contact is also vibration in the spring washer is partially damped. Part of worsened. All these can lead to incorrect wheel guidance on the vibration spreads and continues through the sleeper to the wing rail and can accelerate aging, material degradation the gravel bed. The energy generated by impact and surface in construction and GPK disintegration [6]. roughness of the wheel-rail surface result in shock and Vibrations spread from the source to the structure, vibrations being passed into the entire superstructure and in particular in the longitudinal and vertical directions. At the ballast bed. the same time, transverse vibrations due to the sinusoidal In the rail bed, the gravel grains vibrate, which leads movement of the wheelset in the rail are also passed into to abrasion. Due to abrasion, the rail bed becomes clogged, the frog based on the position of the wheel set entrance. the bearing capacity is reduced and the support of the Piezoelectric acceleration sensors were chosen to capture

COMMUNICATIONS 2/2020 VOLUME 22 THE CONTINUOUS MONITORING OF SELECTED RAILWAY STRUCTURES USING THE AUTONOMOUS... 93

Figure 5 Location of acceleration sensors the description of vibrations spreading. Piezoelectric sensors 5 The vibrations assessment methodology belong among the most commonly used sensors for these measurements. Advantages of these sensors are stability, The test measurement was performed at both positions wide frequency range and power supply independence. during the two working days in May 2019 and included Location of the accelerometers was designed on the base train sets in passenger, express and freight train categories. of the wing rail, on the sleeper under the frog tip and in the Two developed data loggers were used for measurements. gravel bed. A three-axis sensor was designed and placed Measurements were performed under the same climatic on the heel of the wing rail, which allowed monitoring the conditions. The time course of vibration acceleration was dynamic shock caused to the tip of the frog in the vertical the quantity recorded. The start and data storage were direction, transmitted to the foot of the wing rail. accomplished at the pre-set level of a signal detected by the This sensor records vibrations caused by the movement accelerometer placed on a wing rail. Data was transferred characteristic of the wheelset in the longitudinal and to a main computer and evaluated daily. Train sets passing transverse directions. The movement of the wheelset is through points were recognized in a main computer by described as a sinusoid and affects the magnitude of the analysing the acceleration measured in a wing rail. A lateral impact of the phase of movement in which the support-vector machine was used to recognize train sets. It wheelset enters the frog. The next site under investigation is a very promising classification method allowing the train was the transition from the foot of the wing rail to the sets to be recognized by an acceleration signal measured sleeper under the frog tip. A uniaxial accelerometer sensor in a wing rail by using derived characteristics, such as the was used to detect vibrations on the sleeper. The same floating RMS, the number of local maxima, deviations, etc. was true of monitoring the propagation of vibrations into Besides, it is also quite convenient to put together a database the gravel bed. To measure the vibrations in the gravel of locomotives, coaches, and/or whole train sets and their bed, the accelerometer sensor was mounted on a special wheel base for the comparison purposes. A classifier had to hemisphere implemented in the gravel layer [7]. be set before the measurement itself. The speed of passing The designed data logger was used for measurement train sets was calculated from signals derived from the two of the selected dynamic parameters of both switches for accelerometer data recorders (a reference recorder and subsequent comparison and evaluation of the effect of a recorder placed in a wing rail) and the known distance application of the spring-loaded fastening in the switch on between them. A method for determining the speed of train the vibrations spreading of the switch structures. Note that sets is based on measuring the time of a wheel passing for this contribution, the transmission of vibrations from gradually through two acceleration recorders. the wing rail near the frog tip through the sleeper to the It should be mentioned that vibrations transfer from a gravel bed was measured (Figure 5). The set of sensors wing rail in the vicinity of the crossing frog point through a was supplemented with an auxiliary accelerometer, which sleeper to a ballast bed was measured for this contribution. was placed 2 m from the frog tip for easy determination of In total, 3 measuring points (Figure 5) were fixed on each travel speed. switch on which the acceleration sensors (wing rail, sleeper near the fastening and gravel bed near the fastening) were attached. The gravel bed vibrations were measured by an acceleration sensor placed in a special hemisphere pad. This was embedded in the gravel bed in the sleeper crib. Along with the measurement of vibrational characteristics,

VOLUME 22 COMMUNICATIONS 2/2020 94 JANOSTIK et al.

Table 1 Comparison of the effective acceleration values of both monitored turnout structures

-2 SWITCH V EFFECTIVE VALUES OF VIBRATION ACCELERATION [m·s ] Train NUMBER -1 [km·h ] KKz Pz SLz 3 130 52.3 8.9 3.7 RAILJET 4 130 38.2 15.0 7.4

3 146 64.4 9.8 2.8 PENDOLINO 4 160 51.5 13.3 5.9

3 132 52.3 8.0 2.8 LEO EXPRESS 4 130 36.8 14.1 4.8

3 131 53.9 10.7 3.7 REGIOJET 4 127 43.4 16.6 6.3

3 120 53.9 9.8 2.8 R 361 4 120 31.6 11.6 7.2 KKz wing rail, vertical direction Pz sleeper - main branch at frog, vertical direction SLz gravel ballast, vertical direction

3 -jf22r t - a large number of other parameters were measured. Their ASa fa= te$$dt ms$ , (2) ^^hh# 6 @ comparison and analysis are not parts of this paper. -3 For evaluating vibrations during the passage of train sets where f is the frequency, t time, and a(t) the acceleration around a stationary measuring station, it is advantageous to course in the time domain and ASa(f) its representation in use the so-called effective value of acceleration aef or the the frequency domain, j is the imaginary unit. effective value of acceleration value Laef. This quantity best describes the vibrational behaviour of the passing train. The spreading of vibrational waves from moving trains 6 The measurement assessment is characterized by the dependence of the values of the vibration acceleration on the frequency. Therefore, the From the measured records, the basic vibration frequency analysis was also used in the evaluation. characteristics, i.e. the effective value of the acceleration

After analysis of the issue, monitoring measurements value aef, were calculated for each switch and each position and calculations, the following methods and parameters of the acceleration sensor, and amplitude spectra for each were used to analyse the measured data: category of the train. From the calculated data of the • Time waveform of acceleration of the vibration effective vertical acceleration value and the time course • Levels of the effective acceleration value of acceleration, a table was created for both investigated • Frequency analysis using the amplitude spectrum of structures and basic graphs and these were compared. Note the linear axis that the calculated values given in the table were averaged

The mathematical definitions of the individual for each category of the train. The comparison of Laef values parameters used to compare the two constructions are for both measured structures is presented in Table 1. Table briefly summarized. The effective acceleration value is 1 shows the following: defined by the equation [8] The RailJet set has a higher effective vertical acceleration value in the vertical direction for the frog of T 1 2 -2 aef = $$# atdt ms$ , (1) the switch No. 3. The expected difference is in the effective T 0 ^h 6 @ values on the sleepers, the vibration dampening is higher on where a(t) is the instantaneous acceleration value and the frog at the switch No. 3. The values in the gravel bed are T is the time for which the effective acceleration value, slightly lower for the switch No. 3. i.e. the passage time of the train set before the tested For Pendolino, the effective vertical acceleration values construction, is to be determined. are, according to expectations, the greatest compared to The acceleration frequency spectra were calculated by the other kits. Despite different speeds, the acceleration applying the Fourier transform according to the equation values of the individual sensors are in a similar ratio to the [8-9]:

COMMUNICATIONS 2/2020 VOLUME 22 THE CONTINUOUS MONITORING OF SELECTED RAILWAY STRUCTURES USING THE AUTONOMOUS... 95

Figure 6 Time and frequency characteristics of switch No. 3, RailJet, speed 130 km·h-1

Figure 7 Time and frequency characteristics of switch No. 4, RailJet, speed 130 km·h-1

RailJet. The same finding applies to Leo Express, RegioJet mounting node and the graphs in Figure 7 show the outputs and 361 series locomotive sets. of a switch with a flexed mounting node when a high-speed In addition, in this case it is important train travels at a speed of 130 km·h-1. Note that the time and that values in the rail bed are lower for the switch frequency characteristics of vibration measurements are No. 3, i.e. the switch with a sprung fastening against the represented by six graphs in both figures. The graphs on switch No. 4, i.e. with the classical fastening construction. the left represent the measured waveforms of the vibrations Note that the attenuation in the track bed is the lowest acceleration at each acceleration sensor. The graphs on the when the RegioJet was passing through. right represent the calculated amplitude frequency spectra. Table 1 shows that there is a positive effect of the These graphs show several differences between the sprung fastening in the switch No. 3. Due to its use, there two structures. The acceleration values on the wing rail are is a higher attenuation of vibrations passing from the higher in the switch No. 3 compared to the switch No. 4. wing rail to the gravel bed. Analysis and comparison of The same applies to the amplitude spectra. Only for switch time records and their amplitude frequency spectra also No. 3, maximum spectrum values occur between 300 Hz show similar conclusions. This is evident from the graphs and 400 Hz compared to Switch No. 4, where peaks are in Figures 6 and 7. The graphs in Figure 6 show the time achieved at two frequency intervals (60 Hz to 70 Hz, 200 Hz and frequency characteristics of a switch with a classic to 300 Hz). On the sleeper, the lower values were obtained

VOLUME 22 COMMUNICATIONS 2/2020 96 JANOSTIK et al.

in the switch No. 3 compared to the switch No. 4. This also vibration acceleration values exhibit greater damping than applies to the frequency spectra. The significant frequency the frog classically imposed. range here is from about 40 Hz to 120 Hz. Based on the work, the authors tested whether the Likewise, the lower values of the switch No. 3 in the proposed methodology, including the equipment used, ballast bed were achieved. This applies to both time and was satisfactory. It is sufficient to measure such a range. frequency domains. It represents a suitable complementary method of measurement to diagnose and monitoring the behaviour of the structure. At the same time, the team recommends the 7 Conclusions use of measuring stones for further measurements, which have a more suitable location and measure the dynamic A simple diagnostic system developed at the authors’ effects under the tip of the frog in the gravel bed. Note workplace was presented in the paper. It was designed to be that those were also developed at the Institute of Railway used not only to capture basic dynamic data within railway Structures [10]. structures. This system can become a basis for a wider For more accurate evaluation, it is necessary to carry implementation of the concept of intelligent sensing of out measurements more often, preferably continuously selected physical quantities on railway (switch) structures using measuring devices built into the measured sections and their processing in the form of intelligent diagnostics. and structures, or more often so that statistical analysis In addition, such a system will find use not only in can be included in evaluation. A very interesting idea could the field of long-term monitoring of test sections. Based be the installation of sensors in the production of switch on experience with the presented measuring system, the sleepers located under the frog, where the sensor would team recommends continuing in its development. Certainly, measure directly on the contact surface of the sleeper and it would be appropriate to complement this system with gravel. additional sensors and transducers. It would be very beneficial to supplement the proposed equipment with the possibility of measuring the travelled load. A very Acknowledgments interesting application area would certainly be integration of the designed data logger into the diagnostic system, This paper has been supported by the research project which would automatically evaluate the state of the The “Switch and Crossing Optimal Design and Evaluation“ switches, including the occurrence of failures and defects. (S-CODE), supported by the Shift2Rail joint undertaking From the results, it can be stated that evaluation as a part of the Horizon2020 European research program, confirmed the elastic behaviour in the frog seating inthe focuses on the technological vision for the future switch No. 3. When moving from wing rail to sleeper, the generations of switches, under grant agreement No 730849.

References

[1] VEIT, P., MARSCHNIG, S. Towards a more sustainable track. Railway Gazette International. 2011, January, p. 42-44. ISSN 0373-5346. [2] AUER, F. The influence of elastic components on the track behaviour. OVG Tagung Salzburg. 2012, 104, p. 53-55. [3] MORAVCIK, M. Analysis of vehicle bogie effects on track structure-nonstationary analysis of dynamic response. Communications - Scientific Letters of the University of Zilina [online]. 2011, 13(3), p. 33-40. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/860 [4] SMUTNY, J., PAZDERA, L., JANOSTIK, D., NOHAL, V. The monitoring of railway structures using autonomous datalogger. Akustika. 2019, 33(2), p. 81-93. ISSN 1801-9064. [5] SMUTNY J., PAZDERA L., NOHAL V., VUKUSICOVA D. Analysis of vibrations on selected structures of railways. Akustika. 2018, 30(2), p. 74-83. ISSN 1801-9064. [6] SMUTNY, J., PAZDERA, L., NOHAL, V. The analysis of vibrations on selected structures of railways. Akustika. 2019, 31(1), p. 126-135. ISSN 1801-9064. [7] SMUTNY, J., NOHAL, V. The vibration analysis in the gravel ballast by measuring stone method. Akustika. 2016, 25(1), p. 22-28. ISSN 1801-9064. [8] SMUTNY, J., PAZDERA, L. The experimental analysis of dynamic processes related to railway transport. Brno: CERM, 2012. ISBN 978-80-7204-827-4. [9] PAZDERA, L., SMUTNY, J. Evaluation of the railway substructure in the turnout with the movable frog by help of decomposition. In: Applied Methods of the Analysis of Static and Dynamic Loads of Structures and Machines II. Switzerland: Trans Tech Publications, 2016. ISSN 1660-9336, ISBN 978-3-03835-531-1. [10] Evaluation of dynamic effects acting on switches. Certified methodology, 124/2014-710-VV/1, 2014.

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.97-106 97

Romi Ilham - Nanang Shonhadji - Hariadi Yutanto - Diah Ekaningtyas ANALYSIS OF THE ACCEPTANCE FACTOR OF ANDROID-BASED PARKING INFORMATION SYSTEMS IN INDONESIA

Parking is an unrelenting problem, with more vehicles having an impact on the way how to park a vehicle. Some experts have made various breakthroughs in overcoming parking problems; one of them is using smartphone technology as a system to facilitate the way to park. This study aims to analyse user acceptance using the Unified Theory of Acceptance and Use of Technology UTAUT method with structural equation models (SEM-PLS), against 221 respondents. The result is that performance expectations, effort expectancy, and social influence variables have a significant effect on behavioural intention. Then the facilitating conditions variable and behavioural intention significantly influence the use of behaviour for using smartphone parking systems. Furthermore, this research is expected to help the government to find out what factors affect the parking system. Keywords: parking, android, smartphone, UTAUT

1 Introduction smartphone technology parking systems with the Unified Theory of Acceptance and Use of Technology (UTAUT) The problem of parking is classic in big cities, because Model, to determine application performance, ease of it has occurred since many years ago. The problem is that use, social influence and to determine user intentions and because the parking fee is considered the only potential technology acceptance [4]. source of income for regional government development financing [1]. Therefore, in this case, the government is competing to find ways to increase the source of 2 Literature review income through parking by improving the existing system. Furthermore, the most important thing is a great system Indonesia is a developing country; currently, major and infrastructure so that the parking process can run cities in Indonesia have implemented a variety of parking smoothly [2]. systems as a way to increase its regional income sources. In previous studies, Android-based parking systems have According to its development, there are several types of been developed to prevent queues and cost-effectiveness parking systems that have been developed in Indonesia, [2-3]. However, in its journey, it required socialization and and the first is a desktop-based application system This education to the community in its application. There are application can run by itself or independently without various models of parking systems that have been made using a browser or internet connection on a computer by experts. Some using personal computers (desktop PCs) autonomously, with a specific operating system or platform and other use smartphones. Along with the changing times, [5]. The second type of application is web-based; this the use of desktop-pc is considered to be lagging behind application can run using a web technology or browser. This because besides that, it has to take up space and it is not application can be accessed anywhere as long as there is a easy to move. supportive internet connection without the need to install The use of desktop-pc is only suitable for buildings, on each computer, such as a desktop application, only schools, or private companies in a limited scope. The next by opening a browser and going to where the application parking system is the Parking Meter, commonly called server is installed. “ATM Parking” in Indonesia. This system is considered quite The third type of application is mobile-based, this useful because it does not require parking attendants in its application is at a glance almost the same as web-based, application. However, many people do not use Parking ATMs but there are four differences when viewed from features, because there are no officers to guard [3]. Furthermore, user interaction, location awareness, and push notification. the application of the parking system uses smartphone While viewed from application connections, there are two technology Thus latest technology offers digital payment types of connections, namely online connections, which facilities, besides it is easily carried by parking attendants, means we are currently connected to the internet or so officers can adjust the parking area while parking the cyberspace, connected to it through social media accounts, vehicle. This study aims to analyse the acceptance of e-mails, and other types of accounts that we use or use

Romi Ilham*, Nanang Shonhadji, Hariadi Yutanto, Diah Ekaningtyas Department of Accounting, Faculty of Economics, Sekolah Tinggi Ilmu Ekonomi Perbanas Surabaya, Indonesia *E-mail of corresponding author: [email protected]

Figure 1 UTAUT model framework [12] via the internet. While offline is a term for a situation we travel movements must provide adequate service facilities. are not connected to the internet; more precisely it is Increasing population and increasing vehicle ownership not connected. In connection with the application to be will lead to increased demand for roads to accommodate analysed, the analysis is a mobile-based application with an traffic activities [11]. Based on the literature review above, online connection [6]. the following framework is as presented on Figure 1. According to Regional Regulation No. 9 of 2001 Performance Expectancy - as the degree to which concerning Parking Taxes Parking is a non-moving a person trusts by using the system - will help that person to condition of a vehicle that is not temporary. Parking Tax, gain performance gains on the job. In this concept there is from now on referred to as Tax, is a tax imposed on the a combination of variables obtained from previous research operation of Parking Spaces. Parking space is an off-street models about the acceptance and use of technology models parking space provided by individuals or entities, whether [4]. The variables are: perceived usefulness [13], extrinsic provided in connection with the principal of the business or motivation [13], job fit [14], relative advantage [15] and provided as a business, including the provision of Motorized outcome expectations [4]. Based on empirical evidence, Vehicle Storage and Garage of Motorized vehicles that they are separated into performance expectations and collect fees. [7] As it was revealed that roadside parking is personal expectations [13, 16]. Usefulness - a level where managed directly by the City Government while off-street is someone believes that the use of a particular subject - managed by the private sector for public use. will be able to improve the work performance of that The difference between Parking Tax and Parking person. From some of the explanations presented above, Levies on the edge of public roads from now on, referred to it can be concluded that someone trusts and feels using an as levies, lies in the use of the parking space. In levies are information technology will be very useful and can improve levied on payments for the use of public roadside parking performance and work performance [17]. spaces, which are still government-owned facilities. As Effort expectancy is the level of ease of use of the determined by the Mayor or Regent, Parking Taxes are system that will be able to reduce the efforts (energy imposed on payments for the operation of parking spaces and time) of individuals in doing their work [4]. These outside the road body, which means privately owned variables are formulated based on three constructs in facilities and usually managed by a private agency. The the previous model or theory, namely Perceived Ease of potential increase in local revenue through parking is high Use (PEOU) of the Technology Acceptance Model (TAM) because one of the sources of regional income is from fees model, complexity of the Model of PC Utilisation (MPCU), and parking taxes [8]. At present, parking management and ease of use of Innovation Diffusion Theory (IDT) [4]. on public roads is mostly done by the private sector in Davis, et al. [13] identified that ease of use influence the use the name of associations or community groups, so that of information technology. The ease of use of information the parking fund does not contribute positively to local technology will cause a feeling in a person that the system revenue. has a use and, therefore, creates a sense of comfort when The potential income from parking can change along working with it [18]. Several indicators of the ease of use with the business progress of the community [2, 9-10]. of Information Technology (IT), namely: IT is elementary For example, the addition of a parking space affects to understand, IT do easily what is desired by its users, the potential increase in parking revenue; Besides that, user skills will be increased by using IT, and IT is very easy location and usability also affect the income. There are to operate. From some of the explanations given above, three main elements in the transportation system, namely information technology users believe that information vehicles, lane infrastructure, and terminals. The traffic technology that is more flexible, easy to understand, and goes to a destination, and after reaching that place the easy to operate will generate interest in using information vehicle needs a stop. The stop is then called a parking technology and so will use information technology [4]. space. For a vehicle transportation system to be more Social Influence is defined as the extent to which efficient, places that are considered capable of arousing an individual perceives the interests that are trusted by

COMMUNICATIONS 2/2020 VOLUME 22 ANALYSIS OF THE ACCEPTANCE FACTOR OF ANDROID-BASED PARKING INFORMATION SYSTEMS... 99

Table 1 Construct performance expectations

Construct Definition Item Perceived Usefulness [13] The degree to which a person believes that using Using a mobile-based parkingapplication a particular system would enhance his or her job system will increase the effectiveness of my performance. activities. Extrinsic motivation [13] The perception that users will want to perform an - activity because it is perceived to be instrumental in achieving valued outcomes that are distinct from the activity itself, such as improved job performance, pay, or promotion. Job-fit [22] How the capabilities of a system enhance an The use of a mobile-based parking application individual’s job performance. system can reduce the time needed to work. Relative Advantage [15] The degree to which using innovation is Using a mobile-based parking application perceived as being better than using its system can increase my productivity. precursor).

Perceived Usefulness [13] Outcome Expectations related to the With a mobile-based parking application consequences of behavior. Based on empirical system will improve the quality of the results evidence, they are separated into performance of parking activities and will increase the expectations and personal expectations quantity for the same amount of effort.

Table 2 Construct effort expectations Construct Definition Item Perceived Ease of Use [4] The degree to which a person believes that using Learning to operate a mobile-based parking a system would be free of effort. system is easy for me. It would be easy for me to become skilled in using a mobile-based parking system. Complexity [4, 22] The degree to which a system is perceived as Using a mobile-based parking system involves relatively difficult to understand and use. very little time to carry out mechanical operations (e.g., Input vehicle license number data). Ease of Use [4, 15] The degree to which using innovation is Interacting in operating a mobile-based perceived as being difficult to use. parking system is very easy for me.

Table 3 Construct Social Influence Construct Definition Item Subjective Norm The person’s perception that most people who are People who influence my behavior think that I [12, 14, 22] important to him think he should or should not should use a mobile-based parking system. perform the behavior in question.

Social Factors [23] The individual’s internalization of the reference The management of this business has been group’s subjective culture and specific helpful in the use of a mobile-based parking interpersonal agreements that the individual has system. made with others in a specific social situation. In general, the organization has supported the use of a mobile-based parking system. Image [16] The degree to which the use of an innovation is People in my organization who use a mobile- perceived to enhance one’s image or status in based parking system have more prestige and one’s social system. have a high profile then those who do not

others who will influence him using a new system [17]. of information technology will increase the status (image) Social influence is a determining factor for behavioural of a person in the social system. Social influence has an goals in using information technology which is represented impact on individual behaviour through three mechanisms, as subjective norms in the Theory of Reasoned Action compliance, internalization, and identification [4, 19-20]. (TRA), TAM, Theory of Planned Behaviour (TPB), social Facilitating Conditions are defined as the extent to factors in MPCU, as well as images in IDT [4]. Moore and which a person believes that organizational and technical Benbasat [15] state that in specific environments, the use infrastructure is available to support the system. In this

VOLUME 22 COMMUNICATIONS 2/2020 100 ILHAM et al.

Table 4 Construct Facilitating Conditions Construct Definition Item

Perceived Behavioral Control Reflects perceptions of internal and external I have the resources necessary to use a mobile- [5, 22] constraints on behavior and encompasses self- based parking system. efficacy, resource facilitating conditions, and I have the knowledge necessary to use a technology facilitating conditions. mobile-based parking system.

Facilitating Conditions [13] Objective factors in the environment that The guidance was available to me in the observers agree to make an act easy to do selection of the mobile-based parking system. including the provision of computer support). Compatibility [13, 16] (The degree to which an innovation is perceived I think that using the mobile-based parking as being consistent with existing values, needs, system fits into my work style. and experiences of potential adopters).

Table 5 Construct Behavioural Intention Construct Definition Item Behavior Intention The degree to which a person has formulated I wish to use a mobile-based parking system in [4, 11, 24] conscious plans regarding whether to perform a the next three months specified future behavior. I feel benefited from using the mobile-based parking system for the next three months.

Satisfaction [25-26] Favorable intentions to use or acquire the product I will use the mobile-based parking system again or revisit service. when parking my vehicle in the next three months. I will continue to use the mobile-based parking system if the facilities are improved.

Table 6 Construct Use Behaviour Construct Definition Item Use Behaviour Reflects perceptions of the system and potential I often use a mobile-based parking system. [11, 20, 27] future use. I prefer to use a mobile-based parking system

compared to a manual system. Most of my activities are done using smartphones online. Every parking my vehicle always uses a mobile-based parking system.

Table 7 Demographic Respondent Age Gender No Street Name Total 17 - 27 28 - 38 39 - 49 > 49 M F 1 Street Brawijaya 11 25 8 4 35 13 48 2 Street Majapahit 11 7 11 2 27 4 31 3 Street Gadjah Mada 9 15 12 1 33 4 37 4 Street Pahlawan 14 10 13 0 31 6 37 5 Street Raden Wijaya 20 10 11 0 29 12 41 6 Street Empunala 12 6 7 2 21 6 27

concept there is a combination of variables obtained behavioural control [21], facilitating conditions [22], and from previous research models about the acceptance and compatibility [15]. use of technology models. The variables are: perceived

COMMUNICATIONS 2/2020 VOLUME 22 ANALYSIS OF THE ACCEPTANCE FACTOR OF ANDROID-BASED PARKING INFORMATION SYSTEMS... 101

3 Methodology variables. With a limit to reject and accept the hypothesis using a probability of 0.05. See Figure 2 and Table 10. This is an explanatory research. It used a survey The first hypothesis test result is the relationship method to get the data from a particular natural place between the Performance Expectancy variable and to but the researchers do the treatment in collecting data, Behavioural Intention showing a p-value of 0.019 <0.05. for example by distributing questionnaires, tests, and Based on these results it can be concluded that Performance structured interviews [23]. This research was conducted Expectancy has a positive effect on Behavioural Intention using the UTAUT model, a research model that was built (H1 accepted). This result is in line with research conducted to analyse what factors influence the acceptance and use by [30-32], which states that performance expectancy has of technology. The object of this research is the parking a significant effect on behavioural intention. In this case, area of Mojokerto City - East Java. The type of data in the ability of the applied mobile parking application can this study is primary data using field research, which is accommodate the intention to use in terms of the parking research conducted by directly visiting the places that are attendant or the customer. used as research objects. The objects in this study are the The second hypothesis results are the relationship original constructs in the UTAUT model to find out the between the Effort Expectancy variable and Behavioural factors of user acceptance and use of the Ayo application Intention showing the value of p-value <0.001 <0.05. Based in Mojokerto. The indicators to measure each construct in on these results it can be concluded that Effort Expectancy UTAUT are a derivative of the constructs previous studies has a positive effect on Behavioural Intention (H2 is and are presented in Table 1 to Table 6. accepted). According to research from [4, 33], that effort expectancy directly and positively affects the intention. In this digital era, the use of smartphones is very familiar 4 Results and discussion with the public, with an attractive appearance and precise information delivery, making it easier for users to use From the results of a survey conducted obtained data the mobile parking application so that it can lead to the from 221 of 300 targeted questionnaires. The data on the intention to use the application. results of respondents can be seen in Table 7. The third hypothesis is that the relationship between Based on Table 7, the average age of 17-27 dominates Social Influence variables with Behavioural Intention shows more than parking application users of ages 28-38 and the value of p-value <0.001 <0.05. Based on these results the specimen is dominated by male gender, this is in it can be concluded that Social Influence has a positive accordance with a survey conducted by the Association of effect on Behavioural Intention (H3 accepted). A significant Indonesian Internet Service Organizers [28], which states part of the research has proved that social influence has that Generation Z widely uses users of new technology been profoundly affecting human behaviour in general and systems, and men are individuals who are very curious technology adoption in particular [34]. The mobile parking about new things especially smelling of technology [29]. application implemented by the government with the help So that in the future, the mobile-based parking application of advertisements on the media regarding the ease and system is more readily accepted and implemented. benefits of the application can lead to social issues in the The level of validity can be measured by comparing intention to use. the value of r count with the value of r table for a degree The fourth hypothesis is the relationship between the of freedom (df) = n - k with alpha 0.05. If r count is greater Facilitating Conditions variable and the Use Behaviour than r table and the value of r is positive, then the item or showing the value of p-value <0.001 <0.05. Based on these statement is said to be valid. Besides, the validity of the results it can be concluded that the Facilitating Conditions instruments also needs to be tested statistically, by looking have a positive effect on the Use Behaviour (H4 accepted). at the level of significance for each instrument, in this case, There have been many studies that discuss the effect of using Pearson’s total correlation score while the reliability facilitating conditions with use behaviour whose results test uses Cronbach’s alpha, where an instrument is said to have a positive effect, such as research conducted by [4, be reliable or reliable if it has a reliability coefficient of 0.60 18, 32, 35-36], which states that the increasing features of or more. Validity and reliability tests were carried out and an application will have an impact on user behaviour to use tested on 221 respondents randomly. The complete results increasingly. The mobile-based parking application system of testing the validity and reliability are presented in Table is an innovation implemented by the government, has 8 and Table 9. a breakthrough in the form of ease in the parking process, Next is the hypothesis testing stage, which is analysing which is usually managed manually transformed into digital whether there is a significant influence between the automation independent variables on the dependent variable. The path The results of the fifth hypothesis are the relation coefficient can see hypothesis testing, which shows the between variable Behavioural Intention and Use Behaviour, parameter coefficient and the statistical significance value showing the value of p-value <0.001 <0.05. Based on these t. The significance of the estimated parameters can provide results it can be concluded that Behavioural Intention has information about the relationship between the research a positive effect on Use Behaviour (H5 accepted). [21, 37] stated the relation between intention to use and use in

VOLUME 22 COMMUNICATIONS 2/2020 102 ILHAM et al.

Table 8 Validity Variable Dimension Indicator Coeff. Correlation Results PE1 1 0.780 valid PE2 2 0.890 valid Performance Expectancy PE3 3 0.804 valid PE4 4 0.830 valid EE1 1 0.779 valid EE2 2 0.795 valid Effort Expectancy EE3 3 0.784 valid EE4 4 0.791 valid SI1 1 0.805 valid SI2 2 0.836 valid Social Influence SI3 3 0.825 valid SI4 4 0.872 valid FC1 1 0.814 valid FC2 2 0.846 valid Facilitating Conditions FC3 3 0.831 valid FC4 4 0.846 valid BI1 1 0.807 valid BI2 2 0.831 valid Behavioral Intention BI3 3 0.760 valid BI4 4 0.802 valid Age kkp6 1 1.000 valid Gender KS6 1 1.000 valid Experience KS5 1 1.000 valid Voluntariness V 1 1.000 valid KS1 1 0.806 valid KS2 2 0.838 valid Use Behavior KS3 3 0.733 valid KS4 4 0.835 valid

Table 9 Reliability Variable Cronchbach Alpha Results PE 0.896 Reliable EE 0.867 Reliable SI 0.902 Reliable FC 0.902 Reliable BI 0.877 Reliable Age 1.000 Reliable G 1.000 Reliable E 1.000 Reliable V 1.000 Reliable UB 0.879 Reliable

COMMUNICATIONS 2/2020 VOLUME 22 ANALYSIS OF THE ACCEPTANCE FACTOR OF ANDROID-BASED PARKING INFORMATION SYSTEMS... 103

Figure 2 Model analysis using warp PLS Table 10 Summary of results Variable P-value Results Performance Expectancy --> Behavioral Intention 0.019 Accepted Effort Expectancy --> Behavioral Intention <0.001 Accepted Social Influence --> Behavioral Intention <0.001 Accepted Facilitating Conditions --> Use Behavior <0.001 Accepted Behavioral Intention --> Use Behavior <0.001 Accepted social psychology, based on TRA by [38], which states that with behavioural intention with p-value 0.153 <0.05 and the intention to conduct a behaviour is the most significant age variable does not moderate the facilitating conditions predictor of the performance of that behaviours. [39-42] variable with use behaviour with p-value 0.366 <0.05. in his research found that intention to use has a direct The experience variable moderates the social influence positive impact on the use of mobile-based systems, based variable with behavioural intention with a p-value of 0.030 on TAM, researchers such as [4, 27, 43] support the idea <0.05 and moderates the effort expectancy variable with that intention to use has a direct positive influence on the a behavioural intention with a p-value of 0.005 <0.05, use of technology. otherwise the experience variable does not moderate the Furthermore, according to the results that the gender facilitating conditions variable with use behaviour with variable moderates the performance expectancy variable a p-value value 0.372 <0.05. Also, finally, the voluntary with behavioural intention with a p-value of 0.028 <0.05 variable does not moderate the social influence variable and the gender variable moderates the effort expectancy with behavioural intention with a p-value of 0.459 <0.05. variable with a behavioural intention with a p-value of 0.011 <0.05, otherwise the gender variable does not moderate social influence variable with behavioural intention 5 Conclusion, implication and suggestion with p-value 0.382 <0.05. The age variable moderates the performance expectancy variable with behavioural The aim of applied research is to find solutions to intention with p-value <0.001 <0.05 and the age variable improve practice; In accordance with this goal, this study moderates the effort expectancy variable with behavioural adopts the UTAUT model to identify the factors that intention with a p-value value of 0.010 <0.05, otherwise the influence the acceptance of parking services based on mobile age variable does not moderate the social influence variable application systems implemented by the government, and in

VOLUME 22 COMMUNICATIONS 2/2020 104 ILHAM et al.

this case the performance expectations, effort expectations, Monitoring System (CPMMS) technology by mapping social influence, and facilitation conditions are identified available parking slot quotas so as to determine the nearest as factors that influence the intentions and behaviour of parking space, then research conducted by [9] with the mobile parking of government services. Also, the moderate application of Radio Frequency Identification (RFID) role of age, sex, experience, and voluntaries was confirmed. technology that is useful to determine capacity parking The development of cellular telephone network through data collection at the entrance and exit at the technology has now entered its fifth generation, and fast parking area, then there is a study conducted by [46] connectivity is one of the benefits that can be felt by various who applies the number plate recognition, which aims to sectors, one of which is the parking system [44]. Based on recognize the vehicle plate number. this research, it can be seen in the performance expectancy, Research from [7], regarding parking behaviour in effort expectancy, and social influence variables that the user Indonesia, states that many people prefer to park their feels that the presence of a smartphone in a parking system vehicles on the side of the road. From the technology can increase productivity and influence the behavioural that has been done by researchers beforehand, that the intention variable to intend further to use the application application of the technology is not suitable to be applied [4]. Moreover, the facilitating conditions and behavioural on the road because besides the behaviour of Indonesian intention variables influence the use of behaviour, which people who have not been disciplined and structured means that the facilities that support and the intention to parking with the technology is more suitable to be applied use can improve the user’s attitude in using the application by the private sector where there are entrances and exits [4, 15, 38]. such as in malls, school or office building. The impact of this research is felt directly by the local In the future, suggestions for improving the quality government in regulating the parking system, with this and service of the application need to add features system parking management becomes more organized than for government-managed payments such as toll roads, before, and can eliminate the collection of illegal parking, electricity, water, and synergy with government-managed which in turn can increase regional income. Based on payments such as the maintenance of a license (license) the research conducted by [9, 45-46], regarding parking to motor vehicle tax. Parking fees are relatively cheap, but application technology, they argue that the mobile parking parking is one of the factors contributing to revenue for the application in order to be accepted and implemented government [8, 47]. So the government must pay attention must have many facilities and conveniences. [45] conducts to this. research by carrying out Car Parking Management and

References

[1] YUTANTO, H., SHONHADJ, N., ILHAM, R., EKANINGTIAS, D. Development of parking accounting information systems based smartphone in Indonesia. International Journal of Civil Engineering and Technology [online]. 2018, 9(8), p. 1013-1022. ISSN 0976-6308, eISSN 0976 - 6316. Available from: http://www.iaeme.com/ijciet/issues. asp?JType=IJCIET&VType=9&IType=8 [2] HERMAWATI, F. A., KOESDIJARTO, R. A real-time license plate detection system for parking access. Telkomnika. 2010, 8(2), p. 97-106. ISSN 1693-6930. [3] ILHAM, R. Parking management information system based on Android (study case: higher education in Indonesia). International journal of Research Science and Management. 2018, 5(9), p. 1-9. ISSN 2349-5197. [4] VENKATESH, V., THONG, J. Y. L., XU, X. Unified theory of acceptance and use of technology: a synthesis andthe road ahead. Journal of the Association for Information Systems [online]. 2016, 17(5), p. 328-376. eISSN 1536-9323. Available from: http://hdl.handle.net/10397/61599 [5] YOUNG, W., TAYLOR, M. A parking model hierarchy. Transportation [online]. 1991, 18(1), p. 37-58. ISSN 0049-4488, eISSN 1572-9435. Available from: https://doi.org/10.1007/BF00150558 [6] WATENE, G., MUSIEGA, D., NDEGWA, C. A GIS based parking management and dissemination system. International Journal of Science and Research [online]. 2013, 2(7), p. 194-201. eISSN 2319-7064. Available from: https://www.ijsr.net/archive/v2i7/MDIwMTMxMjA=.pdf [7] TEKNOMO, K., HOKAO, H. Parking behavior in central business district. A study case of Surabaya, Indonesia. EASTS Journal. 1997, 2(2), p. 551-570. ISSN-L: 1341-8521, eISSN: 1881-1124. [8] TIMISELA, S., ASNAWI, M., HAFIZRIANDA, Y. Analisis potensi retribusi parkir terhadap pendapatan asli daerah kota prabumulih / Analysis of reception of public parking retribution in the city of Jayapura (in Indonesian). Jurnal Kajian Ekonomi dan Keuangan Daerah. 2004, 2(1), p.1-22. ISSN 2477-7838. [9] IMBIRI, F. A., TARYANA, N., NATALIANA, D. Implementasi sistem perparkiran otomatis dengan menentukan posisi parkir berbasis RFId / Implementation of an automatic parking system by determining the parking position based on RFId (in Indonesian). Jurnal Elkomika. 2017, 4(1), p. 31-46. ISSN 2338-8323, eISSN 2459-9638.

COMMUNICATIONS 2/2020 VOLUME 22 ANALYSIS OF THE ACCEPTANCE FACTOR OF ANDROID-BASED PARKING INFORMATION SYSTEMS... 105

[10] SHIM, S., PARK, S., HONG, S. Parking management system using zigbee. International Journal of Computer Science and Network Security. 2006, 6(9), p. 131-137. ISSN 1738-7906. [11] PATIL, M., SAKORE, R. Smart parking system based on reservation. International Journal of Scientific Engineering and Research [online]. 2014, 2(6), p. 21-26. eISSN 2347-3878. Available from: https://www.ijser.in/archives/v2i6/ SjIwMTMyOTk=.pdf [12] ELWALDA, A., LU, K., ALI, M. Perceived derived attributes of online customer reviews. Computers in Human Behavior [online]. 2016, 56, p. 306-319. ISSN 0747-5632. Available from: https://doi.org/10.1016/j.chb.2015.11.051 [13] DAVIS, A. Failures in adopting green technology under perfect pollution pricing and monopoly. International Review of Economics Education [online]. 2017, 26(March), p. 9-13. ISSN 2347-3878. Available from: https://doi.org/10.1016/j.iree.2017.06.002 [14] SUSSMAN, S. W., SIEGAL, W. S. Informational in uence in organizations: an integrated approach to knowledge adoption. Information Systems Research [online]. 2003, 14(1), p. 47-65. ISSN 1047-7047, eISSN 1526-5536. Available from: https://doi.org/10.1287/isre.14.1.47.14767 [15] MOORE, G. C., BENBASAT, I. Development of an instrument to measure the perceptions of adopting an information technology innovation. Information Systems Research [online]. 1991, 2(3), p. 192-222. eISSN 1526-5536. Available from: https://doi.org/10.1287/isre.2.3.192 [16] COMPEAU, D.R., HIGGINS, C.A. Computer self-efficacy: development of a measure and initial test. MIS Quarterly [online]. 1995, 19(2), p. 189-211. ISSN 0276-7783, eISSN 2162-9730. Available from: https://doi.org/10.2307/249688 [17] HARTINI, S., AFNISARI, K. Unified theory of acceptance and use of technology (UTAUT) on the use of accounting applications on employee performance (in Indonesian). Pilar Nusa Mandiri [online]. 2013, 9(2), p. 113-122. ISSN 1978-1946, eISSN 2527-6514. Available from: https://doi.org/10.33480/pilar.v9i2.134 [18] ZHANG, G., MCADAMS, D. A., SHANKAR, V., DARANI, M. M. Modeling the evolution of system technology performance when component and system technology performances interact: commensalism and amensalism. Technological Forecasting and Social Change [online]. 2017, 125(July), p. 116-24. ISSN 0040-1625. Available from: http://dx.doi.org/10.1016/j.techfore.2017.08.004 [19] HASSANEIN, K., HEAD, M. Manipulating perceived social presence through the web interface and its impact on attitude towards online shopping. International Journal of Human Computer Studies [online]. 2007, 65(8), p. 689-708. ISSN 1071-5819. Available from: https://doi.org/10.1016/j.ijhcs.2006.11.018 [20] HANSEN, J. M., SARIDAKIS, G., BENSON, V. Risk, trust, and the interaction of perceived ease of use and behavioral control in predicting consumers’ use of social media for transactions. Computers in Human Behavior [online]. 2018, 80, p. 197-206. ISSN 0747-5632. Available from: https://doi.org/10.1016/j.chb.2017.11.010 [21] AJZEN, I., FISHBEIN, M. The influence of attitudes on behavior. In: The handbook of attitudes. ALBARRACIN, D., JOHNSON, B. T., ZANNA, M. P. (eds.). 1. ed. Mahwah, NJ: Lawrence Erlbaum Associates, 2005. ISBN 978-0805844931, p. 173-221. [22] THOMPSON, R. L., HIGGINS, C. A., HOWELL, J. M. Personal computing: toward a conceptual model of utilization. MIS Quarterly [online]. 1991, 15(1), p. 125-143. ISSN 0276-7783, eISSN 2162-9730. Available from: https://doi.org/10.2307/249443 [23] NASH, M. S., BRADFORD, D. F. Parametric and nonparametric logistic regressions for prediction of presence / absence of an amphibian. Washington: United States Environmental Protection Agency, 2001, EPA/600/R-01/081. [24] VENKATESH, MORRIS, DAVIS, DAVIS. User acceptance of information technology: toward a unified view. MIS Quarterly [online]. 2003, 27(3), p. 425-478. ISSN 0276-7783, eISSN 2162-9730. Available from: https://doi.org/10.2307/30036540 [25] OLIVER, R. L. Whence consumer loyalty? Journal of Marketing [online]. 1999, 63, p. 33-44. ISSN 0022-2429, eISSN 1547-7185. Available from: https://doi.org/ 10.2307/1252099 [26] ROLPH, E. A., SRINIVASAN, S. S. E-satisfaction and e-loyalty: a contingency framework. Psychology and Marketing [online]. 2003, 20(2), p. 123-138. ISSN 0742-6046, eISSN 1520-6793. Available from: https://doi.org/10.1002/mar.10063 [27] IM, I., HONG, S., KANG, M. S. An international comparison of technology adoption: Testing the UTAUT model. Information and Management [online]. 2011, 48(1), p. 1-8. ISSN 0378-7206. Available from: https://doi.org/10.1016/j.im.2010.09.001 [28] Penetrasi & perilaku pengguna internet Indonesia / Penetration & behavior of Indonesian internet users (in Indonesian) - Asosiasi Penyelenggara Jasa Internet / Indonesia Internet Service Provider Association [online]. 2017. Available from: https://web.kominfo.go.id/sites/default/files/Laporan Survei APJII_2017_v1.3.pdf [29] CARTER, A. J., CROFT, A., LUKAS, D., SANDSTROM, G. M. Correction: women’s visibility in academic seminars: women ask fewer questions than men. PLoS ONE [online]. 2018, 13(9), e0202743. eISSN 1932-6203. Available from: https://doi.org/10.1371/journal.pone.0202743 [30] SAIR, S. A., DANISH, R. Q. Effect of performance expectancy and effort expectancy on the mobile commerce adoption intention through personal innovativeness among Pakistani consumers. Pakistan Journal of Commerce and Social Science. 2018, 12(2), p. 501-520. ISSN 2309-8619.

VOLUME 22 COMMUNICATIONS 2/2020 106 ILHAM et al.

[31] NAHEB, O. A., SUKOHARSONO, E. G., BARIDWAN, Z. The influence of critical factors on the behavior intention to computerized accounting systems (CAS) in cement manufactures in Libya. The International Journal of Accounting and Business Society [online]. 2017, 25(1), p. 86-108. ISSN 2355-2905. Available from: https://doi.org/10.21776/ub.ijabs.2017.25.1.7 [32] PALAU-SAUMELL, R., FORGAS-COLL, S., SANCHEZ-GARCIA, J., ROBRES, E. User acceptance of mobile apps for restaurants: an expanded and extended UTAUT-2. Sustainability [online]. 2019, 11(4), p. 1210. eISSN 2071-1050. Available from: https://doi.org/10.3390/su11041210 [33] GHALANDARI, K. The effect of performance expectancy, effort expectancy, social influence and facilitating conditions on acceptance of e-banking services in Iran: the moderating role of age and gender. Middle-East Journal of Scientific Research [online]. 2012, 12(6), p. 801-807. ISSN 1990-9233. Available from: https://doi.org/10.5829/idosi. mejsr.2012.12.6.2536 [34] GRAF-VLACHY, L., BUHTZ, K. Social influence in technology adoption research: A literature review and research agenda. In: 25th European Conference on Information Systems ECIS 2017 : proceedings. 2017. ISBN 978-989-20-7655-32, p. 331-2351. [35] SHUHAIBER, A. How facilitating conditions impact students’ intention to use virtual lectures? An empirical evidence. In: 12th Advanced International Conference on Telecommunications AICT 2016 : proceedings. 2016. ISBN 978-1-61208-473-2, p. 68-75. [36] ISLAM, M. M. Exploring influencing factors towards intention and use of mobile internet for youth consumers in Bangladesh. Universal Journal of Management [online]. 2017, 5(1), p. 39-47. ISSN 2331-950X, eISSN 2331-9577. Available from: https://doi.org/10.13189/ujm.2017.050105 [37] AJZEN, I., FISHBEIN, M. Understanding attitudes and predicting social behavior. N.J., Prentice-Hall: Englewood Cliffs. 1980. ISBN 0-13-936435-8. [38] DAVIS, F. D. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly [online]. 1989, 13(3), p. 319-340. ISSN 0276-7783, eISSN 2162-9730. Available from: https://doi.org/10.2307/249008 [39] OH, S., LEHTO, X. Y., PARK, J. Travelers’ intent to use mobile technologies as a function of effort and performance expectancy. Journal of Hospitality and Leisure Marketing [online]. 2009, 18(8), p. 765-781. ISSN 1936-8623, eISSN 1936-8631. Available from: https://doi.org/10.1080/19368620903235795 [40] PARK, S. Y., NAM, M. W., CHA, S. B. University students’ behavioral intention to use mobile learning: evaluating the technology acceptance model. British Journal of Educational Technology [online]. 2012, 43(4), p. 592-605. ISSN 0007-1013, eISSN 1467-8535. Available from: https://doi.org/10.1111/j.1467-8535.2011.01229.x [41] SATHYE, S., PRASAD, B., SHARMA, D., SHARMA, P., SATHYE, M. Factors influencing the intention to use of mobile value-added services by women-owned microenterprises in Fiji. Electronic Journal of Information Systems in Developing Countries [online]. 2018, 84(2), p. 1-10. ISSN 1681-4835. Available from: https://doi.org/10.1002/isd2.12016 [42] MUNOZ-LEIVA, F., CLIMENT-CLIMENT, S., LIEBANA-CABANILLAS, F. Determinants of intention to use the mobile banking apps: An extension of the classic TAM model. Spanish Journal of Marketing – ESIC [online]. 2017, 21(1), p. 25-38. ISSN 2444-9709. ISSN 2444-9695, eISSN 2444-9709. Available from: http://dx.doi.org/10.1016/j.sjme.2016.12.001 [43] YU, C. S. Factors affecting individuals to adopt mobile banking: empirical evidence from the UTAUT model. Journal of Electronic Commerce Research. 2012, 13(2), p. 105-21. ISSN 1938-9027, eISSN 1526-6133. [44] IYER, L. Android application for vehicle parking system: “park me”. International Journal of Innovations and Advancements in Computer Science IJIACS. 2014, 3(3), p. 1-7. ISSN 2347-8616. [45] MOSES, N., CHINCHOLKAR, Y. D. Smart parking system for monitoring vacant parking. International Journal of Advanced Research in Computer and Communication Engineering [online]. 2016, 5(6), p. 717-720. ISSN 2319-5940, eISSN 2278-1021. Available from: https://ijarcce.com/upload/2016/june-16/IJARCCE%20159.pdf [46] RASHID, M. M., MUSA, A., RAHMAN, M. A., FARAHANA, N., FARHANA, A. automatic parking management system and parking fee collection based on number plate recognition. International Journal of Machine Learning and Computing [online]. 2012, 2(2), p. 93-98. ISSN 2010-3700. Available from: https://doi.org/10.7763/IJMLC.2012.V2.95 [47] IRFANSYAH, S. Analisis potensi penerimaan retribusi parkir di wilayah kota tangerang selatan / Analysis of the potential for receiving parking fees in the southern Tangerang city area (in Indonesian). Jakarta: Universitas Syarif Hidayatullah; 2018.

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.107-114 107

Mariusz Kostrzewski - Jozef Gnap - Pavol Varjan - Marek Likos APPLICATION OF SIMULATION METHODS FOR STUDY ON AVAILABILITY OF ONE-AISLE MACHINE ORDER PICKING PROCESS

The main aim of the paper is the analysis of simulation model reflecting selected in-warehouse logistics processes in the aspect of their availability. For this purpose, one-aisle machine picking problem with use of a stochastic random simulation is studied, with a special focus on reliability of the system to disturbances and maintenance scheduling. The methodology in the research consists of classic measures of reliability. The model is designed in order to analyze availability of selected parameters of randomly generated order picking process. One of key-results of the paper is answer for question if a mean time to failure can be treated as a value of time when the first failure in the system occurs. A summary of the contribution includes discussion and the perspectives for further research in the subject matter. Keywords: availability, MTTR, MTBF, MTTF, failure rate, failure frequency, warehouse

1 Introduction reduction of traveled distance between picking points, as in [3, 11-12]. Many other authors attempted to reduce OP The order picking (later: OP) process is deliberated process time, as in: [7, 13-14]. as one of the most important among processes in internal In order to achieve as much as realistic simulation logistics. It is believed as such because this process is the model as possible, during its building some events, which most labor consuming one and it determines the level of potentially disrupt a regular material-flow should be service experienced by the down-stream customers [1]. included. This kind of events can be implemented accurately Moreover, it engages most of resources of all processes or randomly. Retooling, maintenance times, accidents, realized in logistics facilities (that aspect is connected to machine failures and many other kinds of disturbances can labor consuming), which is confirmed in [2]. In addition, be included in simulation to make it firstly more realistic, this process is considered to be the most time consuming more adequate to real logistics facility, and secondly [3-4]. All these aspects aim to find out that it is highly to execute what-if analyses in order to be ready for cost-consuming process. Authors of [5] estimated OP cost unpredicted failures or “catastrophes”, which might took - they stated that the cost incurred during OP processes is place in a logistics facility implemented into simulation from 55% to 75% of total cost for all logistics processes in model. In many simulation software there are at least two warehouses (it is also mentioned e.g. in [6-7]). OP process options for failures modeling [15]. The first one is using consists of roughly three phases. The first one is travel to statistical distribution and the second one is application the predefined storage location which is non-value-adding of certain areas based on reliability theory. In the model according to [1], the second one is search of the exact used for this study second option was implemented. By location wanted. This phase can be significant while picking the way, paraphrasing of “failure” meaning given in [16], of small parts, and it is considered also as non-value-adding it is understood in the paper as a unit/machine/mean of phase. The third phase is connected to three actions: reach, transport/device which has not met user/customer/operator take, and deliver to the appropriate in/out spot. This is requirements in some way during its functioning. a value-added and the most resistant to automation activity. Taking into consideration all the above statements, As it was mentioned in many references, the scope of a model for OP analyses in one aisle in OP area was scientific elaborations on OP processes is very broad. The developed (this one aisle is operated by one automated authors deal with e.g.: strategies used to achieve efficient machine: an automated storage and retrieval system AS/RS). OP process (e.g.: [8-10]), cost estimation based on highly The main objective of this model is to obtain a sequence detailed and hierarchical analytical or numerical models, of any number (by default, one hundred) of durations of and reduction of processes duration (or indirectly, by OP process taking place within OP area in a hypothetical

Mariusz Kostrzewski1,*, Jozef Gnap2, Pavol Varjan3, Marek Likos4 1Division of Construction Fundamentals of Transport Equipment, Faculty of Transport, Warsaw University of Technology, Poland 2Department of Road and Urban Transport, Faculty of Operation and Economics of Transport and Communications, University of Zilina, Slovakia 3Office of the Presov Self-governing Region, Presov, Slovakia 4Faculty of Transport, Warsaw University of Technology, Poland *E-mail of corresponding author: [email protected]

warehouse. This paper is connected to the simulation Research results on this subject in the context of model presented in [17] (briefly described in third section logistics (and especially intralogistics) are published much of the paper, the one connected to methodology). For less frequently. Authors of [21] used chosen KPIs in order to this paper, the model would be analyzed in the context of analyze the system composed of a machine, a warehouse, the accessibility of machine contributing to OP process a vehicle and customers. Authors realized it in order to operation and in the context of other concepts related choose a suitable vehicle type according to different costs to mentioned accessibility. Therefore, it can be assumed of this system taking into consideration MTTR and MTBF. that the following parameters are analyzed with use of Authors of [22] prepared a multicriteria rating of the batch the model: availability, mean time to failure, mean time process method, including four KPIs as technological between failures, mean downtime, mean time to diagnose, aspects evaluation in this method. KPIs were also used in mean time to recovery, failure rate, etc. One of the aims of [23] in order to reduce intralogistics costs of spare parts and the article is to answer the following research question. Can semi-finished products while implementation of digitization the simulated times up to the first failure of the means of in maintenance. Authors of [24] used chosen KPIs in order transport and the times obtained on the basis of calculated to analyze buffers next to assembly lines system with statistical values for selected accessibility characteristics material handling. However, several publications touched be treated identically? In other words, can the mean time devices availability analyses for internal logistics, these to failure, calculated from the generalized mean time to were hardly connected to treat it as the main subject matter. failure equation, be treated as the value of time when the first failure in the model occurs? It seems to be important question, especially that a wide range of innovative concepts 3 Methodology, research conditions make transport systems (and equipment) more efficient and and assumptions competitive [18], therefore these systems and equipment need to be analyzed adequately. The methodology consists of several parts. At first, it involves data compilation and pre-processing (despite the model consists of hypothetical data, these must meet 2 Literature review requirements related to the real-world facilities). Secondly, it involves data exploration (mostly, statistics exploration). Several equations for assessing availability of agents Third part includes developing and experimenting with the (e.g. means of transport) taking part in a process can be simulation model, while the last part entails evaluation and found in literature. These equations are based on available interpretation of results of the experiments. data on damage to components over time. These data In modeling theory, the category of models, which are used in order to create an availability model, which selected parameters can be described by random variables, is designed to reproduce as faithfully as possible an distinguishes deterministic and stochastic models. operating process of a machine/vehicle, etc. However, when In a deterministic model none of random variables are hypothetical facilities are analyzed, no historical data is implemented. Stochastic model contains at least one connected to any failures in the process. For this reason, variable of random kind [17, 25-26]. The model considered a simulation modeling may be implemented for analyses. in this paper is stochastic one. Simplifications are adopted for such models, which are The main assumptions of this simulation model are described in methodology section. connected to the fact that it realizes a sequence of randomly The origin of using mean time between failures matched picking lists. The particular picking list is described (MTBF), mean time to recovery (MTTR) was in computer by two parameters. The first one is the number of lines in science and information technology and later in operational each picking list that correspond to assortment of products management. At first, it was used in order to determine to be picked during certain order realization. The second the durability of hard drives. Author of [14] described core parameter is the quantities of items of a certain product to principles of reliability in software engineering. pick from a certain shelf in OP area. The number of lines in As Authors of [19] mentioned, most of companies each picking list (of a j-th element in the sample for k-th implement lean management for their operational experiment) reaches the scope of w(j,k) = {1,…,10}, and the realization. One of the methods, which is connected to lean quantities of i items to pick (per line in a k-th experiment) management is Total Productive Maintenance (TPM). MTBF, are within the scope of p(i,j,k) = {1,…,10}. Picking lists are MTTF, MTTR and OEE (Overall Equipment Effectiveness) generated as a result of initiation of adequate procedure are some of many key performance indicators (KPIs) with pseudorandom number generator (PRGN) included allowing to analyze disturbances and failure rates of in the simulation model, and this PRGN is defined and production processes. In the mentioned paper, it is used constructed by first author based on the logistics map. It for windscreen wipers production, assembly process and starts from random numbers uniformly distributed between internal transport analyses. Meanwhile, Authors of [20] 0 and 1, generated by a suitable standard random number use the term of mean time between overhaul furthermore generator (p*(1,1,k) in Equation (1)), and obtains the sizes - since this parameter is used mostly for engines, it is not of items to be picked using bifurcation of the logistics map considered here. (value 4 in Equation (1) ensures that the data structure is

COMMUNICATIONS 2/2020 VOLUME 22 APPLICATION OF SIMULATION METHODS FOR STUDY ON AVAILABILITY OF ONE-AISLE MACHINE... 109

chaotic), inspired by [27] (in order to obtain integer values [min] divided by the numbers of malfunctions nout (unitless of p(i,j,k), a multiplier by 10 is entered). A picking list in parameter) given according to Equation (3). the model is a series of orders not known in advance that have to be served one after another. More details and PRGN MTTR = Tnouto/ ut (3) verification are given in [28]. MTBF is mean time between failures or disruptions pi,,jk =-10 $$41 pi*,--11jk, - 1 $ in the operation of a product, process, procedure, design, ^ h 6 ^^^ hhh $pi*,--11jk, - 1 (1) machine, unit, mean of transport. Mean time between ^ h@ pk*,11,;! 01 failures assumes that a product - or any other mentioned ^ h ^ entity - can be repaired, and a product can then resume its In order to generate picking lists, logistic differential normal operations [30]. According to [29], the mean time equation is used in the model. It ensures gaining PRGN between failures MTBF is given according to Equation (4) values which are more random than in the case of using in [min]. simply probability distribution. Empirical research with use of simulation model proved that in the case of using chosen MTBF =-T Tnouto/ ut (4) ^h probability distribution results were repeatable in the case of 80% of experiments [17]. In order to exclude repetition, And based on Equations (2)-(4), the availability can be logistic differential equation was introduced. expressed by MTTR and MTBF parameters, as in Equation The simulation model allows to study several (5) - as such is exposed also in [15]. parameters connected to reliability and availability of means of transport used in examined logistics facility. hT =+MTBF/ MTBF MTTR (5) ^h Mentioned parameters are directly connected to reliability or having an impact on it (indirect impact). These MTTF (Mean Time To Failure) determines the average parameters are: processing time, recovery time, cycle operating time of a device from the beginning of its time, capacity, mean time to repair, availability. Moreover, operation or from its last repair to the first failure [31]. This the group of parameters can be broadened by another parameter is particularly important for systems in which ones, which might be useful in a study of reliability and single operations last a long time - small values of MTTF availability of machines and means of transport that are significantly reduces a probability of correct completion not unambiguously defined in the software and require of a single operation. In a lot of design, components and to be redefined in simulation model. These parameters, devices, a value of MTTF is especially near to a value of not exclusively, are mean time to failure (MTTF), mean MTBF (Mean Time Between Failures). Typically, MTBF time between failures (MTBF), estimated percentage of is slightly longer than MTTF [31]. MTTF should be used simulation time during which a mean of transport failure for non-repairable items, however it has been studied may occur (f(k); this is parameter similar to unavailability here in order to examine the interdependence of different - the difference is that f(k) is stochastic parameter, which parameters of this type. value is drawn on the basis of PRGN built into the software In the paper [32] and book [33], the availability is given and its value in relation to availability hT is characterized as Equation (6). by the relation fk 1 1 - hT , whereas unavailability hT ^h Q is characterized by the following relation hhTT=-1 . hT =+MTTF//MTTF MTTR = MTTF MTBF (6) Q ^h The simulation model has been satisfactory verified before and verification proved that the difference between Equation (5) is used in the case, when it is dealing with the results of the simulation model and the analytical model impact of a repairable element on availability of a system is at acceptable level of 3.5%. The model was not validated, (refurbishing/remanufacture is not understood as repair, since it is hypothetical warehouse. but rather replacement). And the Equation (6) is used when According to [29], the availability is equal to the it is dealing with impact of a one-off/non-repairable element probability of a unit/machine/mean of transport when it on availability of a system, in which it operates (an element is operated correctly and without any malfunctions. It is could be refurbished/remanufactured) [30]. The simulation determined by Equation (2), where: hT is parameter of model is assumed to deal with a situation, where replace availability (unitless parameter), T is total operating time parts in equipment subject to failure is possible - therefore

[min] and Tout is a sum of individual periods of downtime Equation (5) is used. [min]. Failure rate [min-1] is the total number of detected defects divided by the total number of samples observed, hTo=-TTut /T (2) and according to [32] is as given in Equation (7). ^h

According to [29], mean time to repair (known also n = 1/MTBF (7) as mean time to restore, mean time to recovery as in [12] or mean downtime, [min]) MTTR is total downtime Tout Failure frequency (unitless parameter), according to [32], is given as in Equation (8).

VOLUME 22 COMMUNICATIONS 2/2020 110 KOSTRZEWSKI et al.

Table 1 Summary of the parameters for assessing the availability of AS/RS Operating f(k) MTTR(k) hT MTTF(k) MTBF(k) n k x k k time ^h Failed [%] ^h [%] [min] [-] [min] [min] -1 [min] [min ] [-] 1 10 1 90 1683.94 8.10 9 0.11111 10.04 169.0675760 2 10 3 90 1737.26 24.30 27 0.03704 10.19 59.0089313 3 10 9 90 1801.22 72.90 81 0.01235 9.20 18.4124711 4 10 27 90 1808.93 218.70 243 0.00412 9.28 6.2173594 5 10 81 90 1682.32 656.10 729 0.00137 8.75 1.8173210 6 10 243 90 1666.35 1968.30 2187 0.00046 14.78 1.0135246 7 20 1 80 1750.16 3.20 4 0.25000 20.17 353.0072720 8 20 3 80 1793.90 9.60 12 0.08333 20.23 120.9686567 9 20 9 80 1914.29 28.80 36 0.02778 21.06 44.7943860 10 20 27 80 1972.89 86.40 108 0.00926 18.70 13.6640900 11 20 81 80 1929.83 259.20 324 0.00309 18.06 4.3028061 12 20 243 80 1666.35 777.60 972 0.00103 20.85 1.4297694 13 30 1 70 1865.32 1.63 2.3 0.42857 30.23 563.8862360 14 30 3 70 1892.27 4.90 7 0.14286 30.87 194.7145830 15 30 9 70 2027.49 14.70 21 0.04762 31.09 70.0385157 16 30 27 70 2054.09 44.10 63 0.01587 26.63 20.2594136 17 30 81 70 2214.97 132.30 189 0.00529 25.92 7.0879040 18 30 243 70 1860.62 396.90 567 0.00176 23.19 1.7756287

x = TMout / TTR (8) calibrating test equipment; dispatching personnel; reviewing technical manuals; complying with supply procedures; and Literature distinguishes three types of availability: awaiting transportation [37]. Operation availability (unitless inherent, operational and achieved. parameter) may be calculated according to Equation (10)

Inherent availability hTI is understood as availability [38]. ^h of a system with respect only to operating time and corrective maintenance. Parameter hTI is equal to hT , hTO =+MTBM/ MTBM MTTM+ LDT (10) ^h given as Equation (5). It does not include standby and delay times and mean logistics delay time (MLDT) [34]. The entire simulation model is predefined, verified

Achieved availability hTA is understood as and realistic thanks to usage of PRGN of two kinds: the ^h availability of a system with respect to operating time predefined in simulation software and the one prepared and both corrective and preventive maintenance (unitless by paper’s first author - specifications of them are given parameter). It ignores mean logistics delay time (MLDT) in the paper. The software predefined PRGN includes and may be calculated according to Equation (9) in [min], MLDT (it was assumed that MLDT is a result of PRGN where MTBM [31] is mean time between maintenance implementation), therefore instead of computing operation in [min] and MMT is mean maintenance time in [min] availability, in this paper inherent availability is computed. [35]. MMT is a measure of maintainability duration - preventive and corrective maintenance are taken into account. It is calculated by adding the preventive and 4 Results discussion corrective maintenance time and dividing it by the number of scheduled and unscheduled maintenance operations Results of k experiments on the model are gathered in during a stated period [36]. Table 1 and Table 2. Experiments allowed to obtain several facts connected to availability, based on values given in hTA =+MTBM/ MTBM MTT (9) Table 1: ^h • with the increase of MTTR(k) ( hT = const) values of

Operational availability hTO is differentiated from operating time oscillate similarly as mean values of OP ^h achieved availability by the fact it includes mean logistics times given in Figures 1-3; delay time (MLDT). MLDT is the indicator of an average time • with the increase of MTTR(k) ( hT = const), MTTF(k) a system is awaiting maintenance and generally includes and MTBF(k) increase exponentially (Figures 4-6) time for locating parts and tools; locating, setting up or

COMMUNICATIONS 2/2020 VOLUME 22 APPLICATION OF SIMULATION METHODS FOR STUDY ON AVAILABILITY OF ONE-AISLE MACHINE... 111

Table 2 Usage of AS/RS, where: mean value of OP process time tk,PS , standard deviation, stk,PS and mean squared ^h ^h error of the mean value stk,PS ^h Operational Operational Failed tk,PS stk,PS stk,PS Failed tk,PS stk,PS stk,PS k Working Waiting ^h ^hk Working Waiting ^h ^h [%] ^h[min] [min] [min] [%] ^h[min] [min] [min] [%] [%] [%] [%] 1 77.54 12.42 10.04 16.84 0.94 0.09 10 63.28 18.02 18.70 19.73 1.16 0.12 2 78.75 11.06 10.19 17.37 1.07 0.11 11 65.00 16.94 18.06 19.30 1.73 0.17 3 80.19 10.61 9.20 18.01 1.01 0.10 12 58.13 21.02 20.85 16.66 0.98 0.10 4 76.25 14.47 9.28 18.09 0.77 0.08 13 47.12 22.65 30.23 18.65 0.53 0.05 5 81.96 9.29 8.75 16.82 0.95 0.10 14 46.33 22.80 30.87 18.92 1.47 0.15 6 67.74 17.48 14.78 16.66 0.98 0.10 15 45.63 23.28 31.09 20.27 2.15 0.22 7 62.38 17.45 20.17 17.50 0.79 0.08 16 51.07 22.30 26.63 20.54 1.80 0.18 8 60.51 19.26 20.23 17.93 0.77 0.08 17 52.40 21.68 25.92 22.15 2.48 0.25 9 60.60 18.34 21.06 19.14 1.57 0.16 18 55.40 21.41 23.19 18.61 2.49 0.25

Table 3 Comparison of MTTF and FTOF (in relation to operating time)

MTTF(k) FTOF(k) MTTF(k) FTOF(k) MTTF(k) FTOF(k) k k k [min] [min] [min] [min] [min] [min] 1 8.10 14.90 7 3.20 138.33 13 1.63 4.25 2 24.30 42.20 8 9.60 41.50 14 4.90 24.90 3 72.90 179.10 9 28.80 220.40 15 14.70 47.56 4 218.70 413.80 10 86.40 379.10 16 44.10 95.10 5 656.10 1180.90 11 259.20 1283.43 17 132.30 275.00 6 1968.30 * 12 777.60 * 18 396.90 1205.70

Figure 1 Mean value of OP process time that accrue Figure 2 Mean value of OP process time that accrue on the on the experiments k = {1,…,6} experiments k = {7,…,12} which confirms the theory and confirms that the c.a. MTTR(k) (MTBF can be understood as a sum of simulation model was verified accordingly; MTTF and MTTR); • 10% reduction in availability results in a halving of • the Figures 1-3 show the duration of OP process MTTF(k) and MTBF(k) values (Figures 4-6); obtained for 18 simulation experiments - the values

• with the increase of MTTR(k) ( hT = const), failure of OP process time are given as average values from rate n k and failure frequency x k decrease 100-elements samples (100 runs of the simulation ^h ^h exponentially, proportional to values of MTTF(k) and model). Figures 1-3 show that the duration of OP MTBF(k), Figures 7-8; process does not increase while MTTR increasing, • failure rate n k and failure frequency x k decrease however changes of durations for consecutive ^h ^h with reduction of availability hT ; experiments can be described by second-degree • as it was mentioned before, MTBF is a little bit longer polynomial functions with determination coefficient, than MTTF - this research proved the theory; and the each given in consecutive figure, for every next difference between MTBF(k) and MTTF(k) is equal to experiment. Nevertheless, increasing of f(k) parameter

and decreasing of availability hT make OP process

VOLUME 22 COMMUNICATIONS 2/2020 112 KOSTRZEWSKI et al.

Figure 6 Comparison of MTTF and MTBF that accrue on the experiments k = {13,…,18} Figure 3 Mean value of OP process time that accrue on the experiments k = {13,…,18}

Figure 4 Comparison of MTTF and MTBF that accrue Figure 7 Changes of failure frequency x k that accrue ^h on the experiments k = {1,…,6} on the experiments k = {1,…,18}

Figure 5 Comparison of MTTF and MTBF that accrue Figure 8 Changes of failure rate n k that accrue on the experiments k = {7,…,12} on the experiments k = {1,…,18}^h

times longer what was expected. In all three cases 5 Conclusion of f(k), the best approximation turns out to be the polynomial of the second degree. All trend diagrams are Availability analysis allows to determine when plotted within the limits range of the standard deviation a machine should be replaced or when a machine should accompanying the bars of the average duration of be passed for technical inspection. In such cases, each OP processes for the individual experiments with component of a machine would have to be treated separately consecutive numbers of k; and not as a whole. For key machines the following principle • MTBF should be as short as possible and it is an overall is generally adopted: preventive replacement of parts every indicator of reliability and effectiveness/efficiency, 85% MTBF [39]. This can be understood that a device should therefore MTTR should be as short as possible. be inspected after this time. At current stage of the model, it can only be concluded that some unspecified failures may Comparison of MTTF and first time of failure FTOF is occur. It is worth noting that in the model, it is not so much given in Table 3. In the case of records with asterisks, the the device’s interference analysis that is important, but the failure did not occur during the experiment; after two days process of internal transport and OP technologies. With this of operation time the simulations were interrupted. assumption, the model is fully functional, and the analyses are satisfactory.

COMMUNICATIONS 2/2020 VOLUME 22 APPLICATION OF SIMULATION METHODS FOR STUDY ON AVAILABILITY OF ONE-AISLE MACHINE... 113

In the introduction section of the paper research is 0.88, therefore the correlation is strong and positive question was given: can the mean time to failure, calculated (two pairs of values were excluded, k = {6, 12} - in their from the generalized mean time to failure equation, be case any failure did not occur during the experiment; after treated as the value of time when the first failure in 2 days of operation time the simulations were stopped). the model occurs? Based on the data obtained during The FTOF values for each k were obtained from individual computation with the model, the answer is “no”, because experiments - in future studies further experiments will be of difference in pair of values MTTF(k) and FTOF(k) given conducted in order to enrich verification of the research in Table 3. However, after computing Pearson correlation question. coefficient for MTTF and FTOF, the value of the coefficient

References

[1] BARTHOLDI, J. J., HACKMAN, S. T. Warehouse & distribution science. Authors own release 0.97. Atlanta, US: Georgia Institute of Technology, 2016 [online]. [Vieved 2018-06-17]. Available from: www.warehouse-science.com [2] ULBRICH, A., GALKA, S., GUNTHNER, W. A. Simulation of multi-level order picking systems within rough planning for decision making. In: SCSC ‘09 Proceedings of the 2009 Summer Computer Simulation Conference: proceedings. Istanbul, Turkey: Society for Modeling and Simulation International, 2009. p. 322-327. [3] LU, W., MCFARLANE, D., GIANNIKAS, V., ZHANG, Q. An algorithm for dynamic order picking in warehouse operations. European Journal of Operational Research [online]. 2016, 248(1), p. 107-122 [accessed 2019-07-17]. ISSN 0377-2217. Available from: https://doi.org/10.1016/j.ejor.2015.06.074 [4] ROODBERGEN, K. J., KOSTER, R. DE. Routing order pickers in a warehouse with a middle aisle. European Journal of Operational Research [online]. 2001, 133(1), p. 32-43 [accessed 2019-07-87]. ISSN 0377-2217. Available from: https://doi.org/10.1016/S0377-2217(00)00177-6 [5] CHIANG, D. M.-H., LIN, C.-P., CHEN, M.-C. The adaptive approach for storage assignment by mining data of warehouse management system for distribution centres. Enterprise Information Systems [online]. 2011, 5(2), p. 219-234 [accessed 2019-07-17]. ISSN 1751-7583. Available from: https://doi.org/10.1080/17517575.2010.537784 [6] DRURY, J. Towards more efficient order picking. 2. ed. IMM Monograph No. 1, Report. Cranfield, U. K.: The Institute of Materials Management, 1988. ISBN 9781870214063. [7] KOSTRZEWSKI, M. Mathematical models of time computing in two-dimensional order picking process in high- bay warehouses. In: Quantitative Methods in Logistics Management. Krakow, Poland: AGH Press, 2014, p. 55-69. ISBN 978-83-7464-713-7 [8] GALAZKA, M., JAKUBIAK, M. Simulation as a method of choosing the order picking concept. Logistics and Transport [online]. 2010, 2(11), 2010, p. 81-88 [accessed 2019-07-17]. ISSN 1734-2015. Available from: https://bit.ly/2Z11j44 [9] KRAJCOVIC, M., GABAJOVA, G., MICIETA, B. Order picking using augmented reality. Communications - Scientific Letters of the University of Zilina [online].2014, 16(3A), 106-111 [accessed 2019-07-17]. ISSN 1338-9777, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/552/517 [10] QUADER, S., CASTILLO-VILLAR, K. K. Design of an enhanced multi-aisle order picking system considering storage assignments and routing heuristics. Robotics and Computer-Integrated Manufacturing [online]. 2018, 50, p. 1-17 [accessed 2019-07-18]. ISSN 0736-5845. Available from: https://doi.org/10.1016/j.rcim.2015.12.009 [11] PAN, J. CH.-H., SHIH, P.-H., WU, M.-H. Storage assignment problem with travel distance and blocking considerations for a picker-to-part order picking system. Computers and Industrial Engineering [online]. 2012, 62(2), p. 527-535 [accessed 2019-07-18]. ISSN 0360-8352. Available from: https://doi.org/10.1016/j.cie.2011.11.001 [12] VENKITASUBRAMONY, R., ADIL, G. K. Design of an order picking warehouse factoring vertical travel and space sharing. The International Journal of Advanced Manufacturing Technology [online]. 2017, 91(5-8), p. 1921-1934 [accessed 2019-07-18]. ISSN 1433-3015. Available from: https://doi.org/10.1007/s00170-016-9879-3 [13] DAVARZANI, H., NORRMAN, A. Toward a relevant agenda for warehousing research: literature review and practitioners’ input. Logistics Research [online]. 2015, 8(1), p. 1-18 [accessed 2019-07-17]. ISSN 1865-035X. Available from: https://doi.org/10.1007/s12159-014-0120-1 [14] TAYLOR-SAKYI, K. Reliability testing strategy. reliability in software engineering [online]. 2016, p. 1-11 [accessed 2019- 07-22]. Available from: https://bit.ly/2MWVBOw [15] BANGSOW, S. Manufacturing Simulation with Plant Simulation and SimTalk. Usage and Programming with Examples and Solutions. Berlin Heidelberg: Springer-Verlag, 2010. ISBN 9783642050732. [16] MUSA, J. D. Software reliability engineering: more reliable software faster and cheaper. 2. ed. Bloomington, Indiana: AuthorHouse, 2004. ISBN 9781418493882. [17] KOSTRZEWSKI, M. Modelowanie i badanie wybranych elementow i obiektow logistycznych z wykorzystaniem metod symulacyjnych / Modeling and research on selected elements and full logistics facilities with the use of simulation methods (in Polish). Warsaw, Poland: Oficyna Wydawnicza Politechniki Warszawskiej, 2018. ISBN 9788378147503

VOLUME 22 COMMUNICATIONS 2/2020 114 KOSTRZEWSKI et al.

[18] SEMENOV, I. N., FILINA-DAWIDOWICZ, L. Topology-based approach to the modernization of transport and logistics systems with hybrid architecture. Part 1. Proof-of-concept study. Archives of Transport [online]. 2017, 43(3), p. 105-124 [accessed 2019-08-12]. ISSN 0866-9546. Available from: https://doi.org/10.5604/01.3001.0010.4229 [19] MICHLOWICZ, E., KARWAT, B. Implementation of total productive maintenance - TPM in an enterprise. Scientific Journals Maritime University of Szczecin [online]. 2010, 24(96), p. 41-47 [accessed 2019-07-18]. ISSN 1733-8670. Available from: https://bit.ly/2KKUpeb [20] WAKIRU, J. et al. Maintenance optimization: application of remanufacturing and repair strategies. Procedia CIRP [online]. 2018, 69, p. 899-904, ISSN 2212-8271. Available from: https://doi.org/10.1016/j.procir.2017.11.008 [21] SADOK, T., NIDHAL, R. Study of a manufacturing system with transport activities in urban area. IFAC- PapersOnLine [online]. 2016, 49(3), p. 419-423 [accessed 2019-07-24]. ISSN 2405-8963. Available from: https://doi.org/10.1016/j.ifacol.2016.07.070 [22] KUKULKA, A., WIRKUS, M. Issues of measuring the course of batch production processes. Procedia Engineering [online]. 2017, 182, p. 387-395 [accessed 2019-07-24]. ISSN 1877-7058. Available from: https://doi.org/10.1016/j.proeng.2017.03.119 [23] FUSKO, M., RAKYTA, M., MANLIG, F. Reducing of intralogistics costs of spare parts and material of implementation digitization in maintenance. Procedia Engineering [online]. 2017, 192, p. 213-218 [accessed 2019-07-24]. ISSN 1877-7058. Available from: https://doi.org/10.1016/j.proeng.2017.06.037 [24] YAN, CH.-B., ZHAO, Q., HUANG, N., XIAO, G., LI, J. Line-side buffer assignment in general assembly line systems with material handling. IFAC Proceedings Volumes [online]. 2009, 42(4), p. 1256-1261 [accessed 2019-07-24]. ISSN 1474-6670. Available from: https://doi.org/10.3182/20090603-3-RU-2001.0310 [25] BUKOWSKI, L. A. Zapewnienie ciaglosci dostaw w zmiennym i niepewnym otoczeniu / Ensuring continuity of supplies in a changing and uncertain environment (in Polish). Dabrowa Gornicza: Wydawnictwo Naukowe Wyzsza Szkola Biznesu w Dabrowie Gorniczej, 2016. ISBN 9788364927508. [26] LESZCZYNSKI, J. Modelowanie systemow i procesow transportowych / Modeling of transport systems and processes (in Polish). Warszawa: Oficyna Wydawnicza Politechniki Warszawskiej, 1999. ISBN 8386569093. [27] GUTENBAUM, J. Modelowanie matematyczne systemow / Mathematical modelling of systems (in Polish). 3. ed. Warsaw, Poland: Akademicka Oficyna Wydawnicza EXIT, 2003. ISBN 8387674532. [28] KOSTRZEWSKI, M. Comparison of the order picking processes duration based on data obtained from the use of pseudorandom number generator. Transportation Research Procedia. 2019, 40, p. 317-324 [accessed 2019-08-12]. ISSN 2352-1465. Available from: https://doi.org/10.1016/j.trpro.2019.07.047 [29] FEDERATION EUROPEENNE DE LA MANUTENTION. Performance data of S/R-machines. Reliability. Availability. 1. ed. FEM 9.221. Frankfort: Germany, 2001. [30] TRIVEDI, K. BOBBIO, A. Reliability and availability engineering: modeling, analysis and applications. Cambridge: Cambridge University Press, 2017. ISBN 9781316163047. [31] STANLEY, S. MTBF, MTTR, MTTF and FIT explanation of terms [online]. [Viewed 2019-07-22]. Available from: https://bit.ly/2rBavBz [32] RHEE, S. J., ISHII, K. Using cost based FMEA to enhance reliability and serviceability. Advanced Engineering Informatics [online]. 2003, 17(3-4), p. 179-188 [accessed 2019-07-17]. ISSN 1474-0346. Available from: https://doi.org/10.1016/j.aei.2004.07.002 [33] BIROLINI, A. Reliability Engineering [online]. Berlin Heidelberg: Springer-Verlag, 2017. ISBN 978-3-662-54208-8. [34] Inherent availability - Glossary of Defense Acquisition [online]. [Viewed 2019-07-23]. Available from: https://bit.ly/31wvr9p [35] Achieved availability - Glossary of Defense Acquisition [online]. [Viewed 2019-07-23]. Available from: https://bit.ly/33mH0l8 [36] MMT - Glossary of Defense Acquisition [online]. [Viewed 2019-07-22]. Available from: https://bit.ly/2Me1Xte [37] MLDT - Glossary of Defense Acquisition [online]. [Viewed 2019-07-23]. Available from: https://bit.ly/2YZLH0X [38] Operation availability - Glossary of Defense Acquisition [online]. [Viewed 2019-07-23]. Available from: https://bit.ly/2KIrgk4 [39] MISIUREK, B. MTBF, MTTR i MTTF - jak i po co stosowac te wskazniki? / MTBF, MTTR and MTTF - how and why use these indicators? (in Polish). Sluzby Utrzymania Ruchu. 2017, 1. ISSN 1896-0677.

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.115-122 115

Boris Zhelenkov - Irina Safonova - Yakov Goldovsky - Elizaveta I. Dmitrieva FORMATION AND OPTIMIZATION OF TRANSPORT NETWORKS IN THE CONDITIONS OF THE DIGITAL INTEGRAL ENVIRONMENT FOR TRANSPORTATION

The article provides a brief description of the digital integrated transportation support environment. The methods of choosing the base route in the formation of transport networks in the conditions of the formation of a digital integrated environment are presented. It is shown that the means of solving the problem of optimization of transport networks is a developed modeling complex, which consists of a set of models with the necessary degree of detail. The paper presents a functional diagram and a brief description of the complex. The complex functions taking into account the replenishment, improvement and updating of models, allow you to combine different types of models and use the necessary of them when solving specific problems of transport network optimization. The modeling complex allows to significantly simplify the optimization of transportation networks. On the basis of the proposed mathematical apparatus, the software of an automated system has been developed, which is designed to function in a digital integrated environment, a brief theoretical description of which is given. Keywords: digital integrated environment, basic route, route criteria and parameters, transport network, optimization, models, modeling complex

1 Introduction Lieberman, Arak, Ligum, Levinson, Black, Vuchic, Schnabel, Lohse, Fernandez and other scientists. The problem is the organization of the information Today there are many mathematical models of space in which both carriers and customers would feel optimization of transport networks [1- 2]. However, all comfortable is very urgent now. To solve this problem, these models are separated and presented in different ways, it is necessary to create a digital integrated environment have a different form of recording, different representation - which is the main component of a unified information of input and output parameters, the same parameters in space to meet the needs of the transportation of passengers different models have different units of measurement, and goods, where user work is positioned as an electronic etc. At the moment there is no single unified form, service. The digital integrated environment should have a a single method of representation of such models, while modular structure that will enable the connection of new individual models do not allow to fully solve the necessary routes, new carriers, as well as representatives of related optimization problems in a digital integrated environment. organizations to provide additional services. When creating Therefore, the development of new more effective models, new routes, it is necessary to evaluate the parameters of approaches and software to optimize transport networks the routes and optimize the movement of route vehicles. applicable in the digital integrated environment is an urgent Consequently, an important step in the development of scientific problem. a digital integrated environment for transportation is the optimization of route transportation networks. The digital integrated environment for transportation 2 Transport network design and optimization is a single information space implemented by software and hardware, which provides the opportunity for interaction A single information space for transportation is a space between carriers and consumers. The digital integrated for interaction between participants in the transportation transport environment is designed to provide information process: employees of transport companies, customers, on all types of transport networks, including urban, regional, authorities and other parties interested in the quality of intercity, personal and freight transport transportation. A single information space for transport The tasks of modeling and optimizing route provision arises as a result of the formation of an appropriate transportation networks were taken up by many specialists. digital integrated environment, including [3-4]: A great contribution to the mathematical formulation of • automated control systems of transport companies; optimization problems and the development of models • clients of transport companies and authorities; of the movement of route vehicles made Varelopulo, • governmental institutions;

Boris Zhelenkov1, Irina Safonova1,*, Yakov Goldovsky1, Elizaveta I. Dmitrieva2 1Department “Computing systems and networks” of the Russian University of Transport (RUT - MIIT), Russia 2Department “Foreign Languages 5” of the Russian University of Transport (RUT - MIIT), Russia *E-mail of corresponding author: [email protected]

Figure 1 Scheme for transport networks design

• corporate and local telecommunication networks, • a system of reference books, classifiers, registers and expert systems, technological databases and reference other regulatory reference information. books; The concept of using such a space involves registering • regulatory databases; a participant in the transportation process on the service • specialized data processing centers that provide timely portal, either as a client or as a representative of the carrier. processing of large volumes of information; The client should be given the opportunity to choose • on-board devices and mobile devices that collect data transportation services based on their preferences. For directly during transportation; example, if cargo is transported, the determining factors • digital communications in the form of global should be the parameters of the cargo, as well as the telecommunication networks, connecting all the listed customer’s preferences - price, speed of delivery, etc. Based components into the information environment. on these and other input data, the route is built. On the way However, the digital integrated environment is not of the cargo, the points of trans-shipment are determined, yet a single information space. A single information space, and the client receives information about the route and in addition to solving other important tasks, requires the the final cost of the service. When implementing passenger development of standards and protocols that ensure a wide transportation, it is necessary to take into account not only exchange of information between its components. To price and time parameters, client preferences for choosing do this, the digital integrated environment must be provided possible intermediate points (cities, attractions), with an for joint work and efficient data exchange between: opportunity of creating a route map with the necessary • automated transport management systems; transport documents, up to booking hotel rooms and • digital interfaces of clients of transport companies; excursions. • a single portal of public services for the access of To implement a digital integrated environment, it is employees and customers of transport companies to necessary to solve a number of problems: development of information; a methodology for creating a digital integrated environment • geographic information systems, including electronic for transport; development of standardization methodology maps, 3D-terrain models and decrypted satellite photos; in the field of digital transportation management; • a unified system of identification, authentication and development of a set of models for the implementation authorization of users of the digital platform; of a digital integrated environment; development of the • payment service to pay for transportation, mutual concept of connecting carrier modules; development of settlements of participants in the transportation an intelligent interactive interface with a digital integrated process, payment of taxes and fees; environment (carrier/customer). In addition to the above, an

COMMUNICATIONS 2/2020 VOLUME 22 FORMATION AND OPTIMIZATION OF TRANSPORT NETWORKS IN THE CONDITIONS OF THE DIGITAL... 117

important task in creating a digital integrated environment of video processing is carried out using innovative for providing transportation is the task of forming and technologies in the field of artificial intelligence. When optimizing transportation networks. A transport network is analyzing the video sequence, additional opportunities arise a set of possible routes of all modes of transport included for the control of emergency situations. The processing in the process of transporting passengers or goods between of information received from video monitoring systems two endpoints. Since several modes of transport can can be performed locally or centrally. Local processing be used for one route, such a system can be called an of information requires the presence of an on-board integrated transport infrastructure or environment. The specialized computer on a vehicle or traffic control means process of creating transport networks is presented as a (traffic lights). The result will be transferred to the dispatch diagram in Figure 1. center. Centralized processing of information reduces the The process of design of transportation networks requirements for local computing facilities, but increases includes three main stages: the load on data transmission channels. At this stage, a set Stage 1. Collection and analysis of initial information. of measures to ensure reliability and information security Information is collected from video monitoring systems is being developed. and DVR systems. They also include video monitoring of The result is a transport network - an electronic map public transport and outdoor security cameras. To collect with complete information about the state of traffic flows, and analyze initial information, the following are needed: emergency situations and the loading of vehicles on routes. systematization and standardization of video recording systems; systematization of information processing. It is necessary for the control system to include the means of 3 Choosing the optimal route for the transport satellite navigation. network Stage 2. Formation of the transport network. Passenger or cargo can be moved between two endpoints using several modes of transport. Each type of 2.1 Choosing the best route for inclusion transport has its own network of routes that are already in the transport network formed and regularly used in various geographical areas. When you include any type of transport in the integrated At this stage it is necessary to make an informed transportation system, all its routes in the corresponding decision about which route should really be included in geographical territory are added to the database and can the transport network, using the method of choosing the be used to build the necessary integral (composite) route. optimal route (it is reasonable to call it a base route) from By integral route we mean a route using various modes a variety of alternatives. of transport. When creating a digital integrated transport Generally, when there is an ambiguity in the choice of support environment, existing routes can be completely a route (usually this occurs under conditions of uncertainty transferred to the digital integrated environment, and new in the initial information), i.e. the only best option cannot routes will be added taking into account the existing ones. be found. You can get only a few («area») rational options. To determine new routes, it is necessary to evaluate Considering this circumstance, one should first identify this the parameters of routes and optimize transport networks. set of rational routes, the final choice of the optimal (base) Estimation of route parameters and optimization of route that will be included in the transport network is made transport networks leads to increased profitability and by applying various expert procedures. With such a final quality of services of transport companies. choice, certain formalized techniques can be applied (for When choosing the best route one should use an example, expert assessment methods, hierarchy analysis appropriate method of selecting the best option from a method), however, the main role belongs to the person - the variety of alternatives. The choice of a route is a multi- expert. criteria task, and many of these criteria are difficult to describe accurately. It is necessary to determine and take into account all the requirements for the route, parameters 2.2 Optimization of transport networks and criteria for its evaluation. Route parameters are - mode of transport, traffic fluctuations, peak loads, risks, cargo The study of modern methods for optimizing transport parameters, infrastructure condition, range, time delays of networks has shown that since the beginning of the the route, tax rates, volume and condition of the fleet of 2000s, new approaches to building and optimizing transport transport companies, quality of delivery and many others. networks have been actively introduced in European Under the conditions of the formation of a digital countries [5]. Existing optimization models should be integrated environment, when choosing a base route to considered here and/or completely new ones should be form a transport network, it is advisable to use interactive developed. decision-making methods, since this allows using the Stage 3. Evaluation of the results obtained is also creative capabilities of a person - decision-maker (DM) carried out by collecting information from video monitoring [6-7]. An expert acts as a decision maker, for example, he systems, from entrance and exit turnstiles, etc. Analysis can be a transport company manager, government official,

VOLUME 22 COMMUNICATIONS 2/2020 118 ZHELENKOV et al.

etc. However, in the process of developing the digital route to be included in the transport network can be integral medium of transportation, there will be a tendency represented as follows: to unite the means of neural networks, expert systems A. Data input. in creating multi-stage decision support systems for the B. Criteria analysis. formation of route networks. C. The definition of a set of rational routes based on The choice of the base route should be made taking the method of analysis of payment matrices with the into account all the factors that determine the quality of selection of the main criterion. such a decision, including technical and organizational, C.1 Selection of representative combinations of source as well as cost, and time ones. This requires the use of data. specific techniques and methods involving the use of C.2 Search and preliminary analysis of solutions. intuition and the decision-maker’s experience. Decisions C.3 Calculation of the payment matrix. made must be steady and regular. Stability of solutions C.4 Analysis of the payment matrix and the choice of should be understood as the property to remain unchanged rational decisions. with respect to changes in a priori probabilities affecting D. Selecting the base route using the hierarchy analysis the values of the estimated functionals. Decision stability method. analysis allows DM to identify solutions with a certain D.1 Assessing the importance of criteria by pairwise guaranteed level. A stable solution is different in that it is comparisons. resistant to DM errors in terms of a priori probabilities. D.2 Calculation of criteria weights. Regularity of decisions means that a complete regular D.3 Comparison of the importance of alternatives by solution has the property that it is optimal for all criteria criteria. considered, but this is extremely rare. In this case, it D.4 Calculate the weights of alternatives for each is advisable to use the set with partial regularity of the criterion. solution, which is optimal only for individual criteria. E. The end of the algorithm. In formal terms, the following decision-making For the analysis of routes, the selection criteria conditions that arise when choosing a base route during proposed for the conditions of uncertainty are used - Wald, the formation of a transportation network [8-9] can be Laplace, Savage, Hurwitz criteria or their combination [13]. distinguished: certain conditions when all the information for forming a base route is precisely known; conditions of uncertainty, when, in addition to unambiguous source 4 Optimization of transport networks based on the data, there are random variables with precisely known modeling complex probabilistic characteristics, as well as quantities for which their probabilistic description is not precisely known or not When modeling transportation in Russia, the software known at all. packages PTV Vision, AIM Sun and others are mainly used. A study of decision making methods showed that for Their main disadvantage is the closed architecture, that conditions of certainty of initial information about routes, is, the impossibility of adding fundamentally new designs it is advisable to use the lexicographic method of organizing (calculation models). Existing modeling tools use only a variety of alternative options when choosing a base separate models and methods that do not allow to fully route [10]. The essence of the method is to streamline the and adequately solve the tasks set for transport workers. decisions in order of criteria importance. In the process of A means of solving the task of optimizing transport choosing a base route from the set of route parameters, networks is a modeling complex, which consists of many you should choose the one that, in the specific application models that have the necessary degree of detail: conditions, reflects the necessary route properties, for example, time delays, and gives it a criterion function. MM= 1,,ffMMis,, , (1) " , As practice shows, the majority of decision-making tasks on choosing a route to include it in the transport with M1, Mi, Ms being specific models. network are solved under conditions of uncertainty of The complex can include both analytical and simulation the initial information. Analysis of existing approaches models for optimizing route networks. The functional to solving the problem of choosing the base route in the diagram of the modeling complex is presented in Figure 2. face of uncertainty of the initial information showed the The fundamental properties of the modeling complex: feasibility of a two-stage decision-making process. At the - property of integration, that is, the complex must first stage, a set of rational routes is determined on the integrate models depending on the specific situation basis of formal methods, on the second, a basic route is (task); adopted on the basis of expert procedures. Effective ones - the property of evolution, which consists in the fact are the methods of analysis of payment matrices with the that the modeling complex, in fact, should be a working allocation of the main criterion and the method of analysis space for the gradual improvement of the transport of hierarchies, respectively [11-12]. network; For the conditions of uncertainty of the initial - the property of duality, reflecting the possibility of information, the general algorithm for choosing the base representing the complex, on the one hand, as a model

COMMUNICATIONS 2/2020 VOLUME 22 FORMATION AND OPTIMIZATION OF TRANSPORT NETWORKS IN THE CONDITIONS OF THE DIGITAL... 119

Figure 2 Functional diagram of the modeling complex

of the object under study, and, on the other hand, as a of large dimensions; effective software implementation; model, reflecting the process itself of optimizing the reliability; convenient visual perception. transport network; In order to overcome the current diversity of the - versatility or applicability for solving an unlimited set representation of the existing models for optimizing of various problems of optimizing transport networks, transportation networks, a single form and a unified taking into account the necessary parameters, criteria methodology for presenting models, their input and output and restrictions; parameters, and units of measurement of parameters have - the property of development, which consists in the fact been developed for the modeling complex. that the complex is created and operates taking into The models of the complex should be presented in the account the replenishment, improvement and updating form of the following dependence: of optimization problems and models of their solutions; the complex allows you to combine different types of & model model number: {input parameters} {output parameters} (2) models and use the necessary ones in solving specific in out problems of optimization of transport networks in a or , Mi | XXpri & pri (3) digital integrated environment of transport. in in in Requirements for the modeling complex: adequacy; with Xpri = xxpri1,,f prin being input parameters for the # - the possibility of refining the model when developing Mi model, the transport network and the possibility of multi-level Xout = xxout ,,f out being output parameters pri " pri1 pril , modeling; correctness of work; the ability to process data (parameter) Mi.

VOLUME 22 COMMUNICATIONS 2/2020 120 ZHELENKOV et al.

Taking into account Equations (2) and (3), Equation (1) when solving a specific task, to establish a quantitative is represented as a system of models as follows: relation between the input and output parameters of the models. The module implements the preparation and ]Z in in in out b_ ]Mx1 | pr11, xxpr12,,ffpr1f & x pr1,;b analysis of model data. ] # - " , b ]h b Results analysis module. According to the results of ] in in in out b M = ][Mxi | pr1,,xxii2 ff,;pr ji& x pr , b` , (4) the analysis of the data obtained during the operation of ] # - " , b ]h b the modeling complex, a message is displayed containing ] in in in out b ]Mxs | prsp12,,xxrs ff,,prsn & x prs . b an assessment of the results obtained and, if necessary, \ " , " , a recommendations for correcting them or adjusting the The output parameters of one model / models can be initial task of optimizing route networks. input parameters for another or other models. The number The modeling complex functions as follows. Based on and types of models are determined by the nature of the the data entered by the user (this may be an optimization problems of optimization of route transport networks. problem, a set of route parameters, conditions and It must be emphasized that Equations (2) and (3) optimization criteria), the input information is analyzed. determine only the form of representation of models, i.e. Next, we determine the necessary optimization tasks for how various models for calculation should be presented in transportation networks (available in the task table or the base of complex models. This is necessary to simplify completely new), the parameters to be obtained; necessary the work with various models of the complex, and to for solving selected problems of the model from the implement the fundamental properties of the modeling database. It is checked whether this initial information complex. is sufficient, if not, then these data are taken from the The composition of the modeling complex. Interface of corresponding tables, database or requested from the user. user allows for problem-oriented communication between Selection and analysis of models (from the database models) the user and the system that implements the operation of is carried out in the required order. Then, the organization the complex. of the interaction of the selected models, calibration models In the databases (DB), the models and parameters (if necessary). Next, the resulting solution is analyzed. As of the models are presented, as well as the optimization a result, a model will be chosen to solve the problem of criteria for transport networks. The tables contain a list of optimizing transport route networks, corresponding to specific optimization problems and statistical information. specified conditions. The module analysis of input data. The functional Depending on the tasks to be solved and the initial purpose of the module is to form a data structure. Data conditions (initial information), various mathematical entry and selection of the optimization task is carried out models and methods that are presented in the model at the user’s request, while reading information from the database, for example, methods based on OD-matrices, corresponding tables and the database. The composition can be used in the calculations. In interactive mode, the of the input data used may vary from one task to another. modeling complex offers the user appropriate models for The module of the organization of the interaction of solving problems, and the user himself can specify the models. Here processes of the organization of interaction necessary models and methods that he will need in the of models are realized. On the basis of existing tables calculations. and databases, statistical processing of information about objects takes place - reduction of the dimension of the feature space, i.e. the selection of such a subsystem of 5 Software for the system of formation features that would be smaller in volume than the prior and optimization of transport networks space of features, and provided an acceptable solution. The analysis of the characteristics of models, analysis On the basis of the proposed mathematical apparatus, and coordination of units of measurement of input and software of an automated system for the formation and output parameters of models in the process of solving each optimization of route transportation networks, designed specific problem of optimization of transport networks, to work in a digital integrated environment, has been for example, the problem of calculating the coefficient of developed. imposing passenger traffic on the transport network [14-15]. The automated system was developed by the staff Module for analysis of relation between parameters of the department “Computing Systems, Networks and models. Functional purpose - determination of the and Information Security” of the Russian University of necessary parameters to solve specific problems and the Transport. Programming language is C ++; OS are Ms. construction of criteria dependence. Windows 8x and above. Calculation module. The values of parameters obtained The software complies with the following basic on the basis of theoretical calculation methods (by criteria) principles necessary for functioning in the conditions can be the initial data for subsequent calculations when of the digital integrated environment for the provision optimizing transportation networks. of transport: systemic unity, development, compatibility Module for assessing the accuracy of models. It is and standardization. The principle of systemic unity in intended to assess the accuracy of each model considered the creation, operation and development of software

COMMUNICATIONS 2/2020 VOLUME 22 FORMATION AND OPTIMIZATION OF TRANSPORT NETWORKS IN THE CONDITIONS OF THE DIGITAL... 121

Figure 3 Software Interface

Table 1 Modelling time reduction with the developed software Modelling Stages Time reduction (%) Information analysis 20 - 30 Estimation of route parameters 10 - 30 Rational options design 10 - 25 Base route selection 15 - 20 should ensure the integrity of the links between system Table 1 shows the reduction in modelling time as components. The principle of development is that the a result of using the developed system. software must be created and function, taking into account As a result of the system operation, an electronic map the replenishment, improvement and updating of its with full information about the optimal (basic) route, the components. The principle of compatibility ensures the formed transport network, and information about the state joint functioning of languages, symbols, codes, information of traffic flows will be obtained. and connections between components and preserve the open structure of the system as a whole. The principle of standardization lies in the unification, characterization and 6 Conclusions standardization of software that is invariant to routes and transport networks. In the practical application of the developed When forming the transport network in the developed mathematical apparatus and the developed software, the system, decision-making methods are implemented both effect is expected in the following areas: under certain and uncertain conditions of the initial • increase the capacity of transport systems, for example, information. The relations between the decision-maker and regional; the computer goes through a series of iterative steps, with • improving the economic efficiency of the use of route two points being significant at each iteration: the computer rolling stock; offers the DM a preliminary decision; the DM expresses • attracting customers by expanding the range of services his opinion about this decision and, if necessary, sets the and customer-oriented approach; direction for finding a solution. Analysis of the proposed • attracting investments from potential partners wishing solution is carried out in a dialogue with a computer. The to provide their services in the digital information route that satisfies the DM is taken as the base route. environment; The system interface is shown in Figure 3. • increasing the level of comfort and safety of route The left screen shows the selection and analysis of the traffic. parameters of the route for inclusion in the route network. Unlike the existing ones, the developed system allows The right screen shows the process of formation and taking into account all the parameters and characteristics optimization of the route transport network. of routes most fully; carry out calculations and processing of data of large dimensions; enables to solve the problems

VOLUME 22 COMMUNICATIONS 2/2020 122 ZHELENKOV et al.

of formation and optimization of route transport networks functioning of the modeling complex, including the concept more adequately and at the lowest cost, allowing to reduce of organizing the interaction of models, the algorithm for the time, while ensuring efficiency of operation. linking model parameters, the integration technique, as well It should be noted that in subsequent articles a more as the stages of the automated system and experimental detailed description of the developed mathematical testing of the developed mathematical apparatus and apparatus will be presented, the composition and System software.

References

[1] KORYAGIN, M. E. Equilibrium models of urban passenger transport system in the conditions of conflict of interests. Kuzbass State Technical University, Novosibirsk: Science, 2011. ISBN 978-5-02-032236-3. [2] GOLDOVSKIY, Y. M., ZHELENKOV, B. V., SAFONOVA, I. E. Simulation of network database management system with limited circulation of segments. Mir Transporta Journal (MT). 2014, V.12, 4(53), p. 32-44. ISSN 1992-3252. [3] SAFRONOV, E. A., SAFRONOV, K. E., SEMENOVA, E. S. Download management of the city transport network, given the increase in the availability of passenger transport. The Russian Automobile and Highway Industry Journal [online]. 2015, 6(46), p. 38-44 [accessed 2019-02-02]. eISSN 2658-5626. Available from: https://vestnik.sibadi.org/jour/ article/view/237?locale=en_US [4] PONOMAREVA, M. S. Coordination of carrier interests in the regional passenger traffic segment. Science studies [online]. 2015, 7(5). [accessed 2019-06-09]. ISSN 2223-5167. Available from: naukovedenie.ru/PDF/151EVN515.pdf [5] KOLBER, Y. A. Information support of public passenger transport network optimization in major cities: relevance and approaches. The Russian Automobile and Highway Industry Journal [online]. 2018, 15(2), p. 218-228. [accessed 2019-06-09]. eISSN 2658-5626. Available from: https://doi.org/10.26518/2071-7296-2018-2-218-228 [6] SMIRNOV, M. I., KHAIROULLIN, R. Z. Mathematical models used in the optimization system of goods delivery by motor transport DISPATCHER [online]. Moscow: Preprint, Inst. Appl. Math., the Russian Academy of Science, 2002. Available from: http://www.keldysh.ru/papers/2002/prep13/prep2002 13.html [7] SHVETSOV, V. I. Mathematical modeling of traffic flows. Automation and Remote Control. 2003, 11, p. 3-16. ISSN 0005-1179. [8] SAFRONOV, E. A., SAFRONOV, K. E., SEMENOVA, E. S. Prospects of development of light rail transportation in Russian cities. The Russian Automobile and Highway Industry Journal [online]. 2016, 2(48), p. 62-70 [accessed 2019- 09-10]. eISSN 2658-5626. Available from: https://vestnik.sibadi.org/jour/article/view/286?locale=en_US [9] FEDOROV, V. A. On the issue of the possibility of optimizing the route network of urban passenger transport in megalopolises. Young Scientist [online]. 2015, 2(82), p. 331-333 [accessed 2019-09-10]. eISSN 2077-8295. Available from: https://moluch.ru/archive/82/14765/ [10] CHERNORUTSKY, I. G. Decision making methods. Science, St. Petersburg: BHV-Petersburg, 2005. ISBN 5-94157-481-9. [11] SAATY, T. L. Making decisions. Hierarchy analysis method. Translated from English by VACHNADZE, R. G. Moscow: Radio and Communication. ISBN 5-256-00443-3. [12] GONCHAROV, V. A. Optimization methods [online]. MIET, Moscow: Jurait, 2016. ISBN 978-5-9916-3643-8. Available from: https://biblio-online.ru/bcode/378897 [13] SAFONOVA, I. E, TELNOV, G. G. The process of traffic control routing of passenger transport in the regions. In: the XIV International Scientific and Practical Conference : proceedings. Vol. 1. Innovative, Information and Communication Technologies. Moscow: Association of Alumni and Employees VVIA them. prof. Zhukovsky, 2017. ISSN 2500-1248, p. 300-302. [14] SAFONOVA, I. E, TELNOV, G. G. Hierarchical modeling complex as a basic means of optimizing route transport networks. In: the XIV International Scientific and Practical Conference : proceedings. Vol. 1. Innovative, Information and Communication Technologies. Moscow: Association of Alumni and Employees VVIA them. prof. Zhukovsky, 2017. ISSN 2500-1248, p. 295-299. [15] SAFONOVA, I. E, TELNOV, G. G. Selecting the optimal passenger transport routes for inclusion in the region route network. In: the XIV International Scientific and Practical Conference : proceedings. Vol. 1. Innovative, Information and Communication Technologies. Moscow: Association of Alumni and Employees VVIA them. prof. Zhukovsky, 2017. ISSN 2500-1248, p. 290-294.

COMMUNICATIONS 2/2020 VOLUME 22 ORIGINAL RESEARCH ARTICLE https://doi.org/10.26552/com.C.2020.2.123-128 123

Martin Boros - Matej Kucera - Andrej Velas - Jan Valouch POSSIBILITIES FOR EXPERIMENTAL TESTING OF ALARM TRANSMISSION SYSTEMS

In the current digital era, information is a basis of some systems. In the area of information, great emphasis is also placed on its security and possibilities of use. The basis of the alarm transmission system is information about the protected object, which is transmitted to the remote center of the alarm transmission system. Operators of alarm transmission system centers should be obliged to carry out regular testing of the availability of individual transmission networks. At present, there is a trend that those tests are carried out by telephone calls between the two technicians and the time of transmission of information is measured utilizing a stopwatch. To automate this process, a test facility has been created that can simulate and record the intrusion of a protected object. Initial experimental tests have ascertained whether it is possible, with the test equipment, to generate the data necessary to assess the reliability of alarm transmission systems. Keywords: test facility, alarm transmission system, information, protected object, alarm transmission system center

1 Introduction by a separate technical standard which, in addition to the functional ones, includes the design requirements, as well. An object protection system can be understood as Given the nature and importance of the alarm a thought-out arrangement of security features, measures transmission system, one could divide it into three basic and their properties that all together create a sense of parts, which are the transmitting device in the protected security. These elements could be divided into four basic building, the transmission path and the receiving device groups: passive protection elements, active protection in the monitoring center. The configuration and setting elements, physical protection elements and regime- of the alarm transmission system shall, depending on the organizational measures. The passive protection elements technical standards, correspond to the logical continuity include mechanical barrier systems, usually supplemented shown in the block diagram in Figure 1. with apertures. The alarm transmission system, together There are several definitions in literature of the system with the electrical alarm system, video surveillance system, reliability, depending on the area in which the system is fire alarm system, access control system and others, belong located. The basic definition, based on technical standards, to the group of alarm systems as active protection elements. defines reliability as the property (ability) of a product or Given the name of the group, one can say that the basis system to perform the desired function within a specified of an alarm system is an alarm, which can be defined as time, while maintaining operating parameters. Reliability a warning of the presence of a danger to life, property or can be expressed by using one of its basic parameters: environment [1-2]. no-failure operation, machine life, reparability, machine An alarm transmission system is defined, according availability [3]. to technical regulations, as an alarm transmission device Based on categorization of the backup transmission and a network used to transmit the status information of paths use, one assumes that the transmission paths one or more alarm systems to one or more alarm devices designated SP1-6 do not use the backup transmission path. of one or more alarm reception centers. In addition to the The DP1-4 paths use an alternative, backup path, depending basic and specific requirements for alarm systems, the on the capabilities of the attached, protected object. technical standards also include the concept of an alarm To increase efficiency and reliability of an alarm transmission system monitoring center, which is understood transmission system, the connection of the object via as a continuous service center in which the condition a backup transmission path is used in practice, which is and functionality of one or more alarm transmission used in the case of a primary malfunction. Due to this systems are monitored. The monitoring center of the possibility of connection of objects, the technical standard alarm transmission system is comprehensively addressed divides them into two basic groups, whose transfer times

Martin Boros1,*, Matej Kucera2, Andrej Velas1, Jan Valouch3 1Department of Security Management, Faculty of Security Engineering, University of Zilina, Slovakia 2Department of Measurement and Applied Electrical Engineering, Faculty of Electrical Engineering and Information Technology, University of Zilina, Slovakia 3Department of Security Engineering, Faculty of Applied Informatics, Tomas Bata University in Zlin, Czech Republic *E-mail of corresponding author: [email protected]

Figure 1 Block diagram of the alarm transmission system

Table 1 Maximum message transmission time and system availability values for weekly and annual periods SP1 SP2 SP3 SP4 SP5 SP6 DP1 DP2 DP3 DP4 Maximum transmission time 480 120 60 60 30 30 120 60 60 30 allowed [s] Weekly period [%] V V V 97.0 99.0 99.8 V 99.0 99.8 99.8 Annual period [%] V V 97.0 99.0 99.5 99.9 V 99.5 99.9 99.9 V = optional could be divided according to the degree of risk. The Reliability of the alarm transmission system is technical standards for alarm transmission systems define a fundamental probability indicator affecting the cumulative four basic levels of risk in terms of the importance of the probability of an object protection system. connected object, where level 1 represents a low risk and The basic indicator of reliability, in terms of alarm level 4 a high risk. From the above and using Table 1, one transmission systems, is their availability. Availability is the can state the following classification: percentage of time that the system or parts of it operate • Level 1: categories SP1, DP1, according to requirements of the technical standards. One • Level 2: categories SP2, DP2, could say that it is an expression of the time during which • Level 3: categories SP3, SP4, DP3, the alarm transmission system can transmit information • Level 4: categories SP5, SP6, DP4. from the protected object to the alarm transmission system Performance of the alarm transmission system, one center without disturbing or devaluing it. The technical of the main indicators of system reliability, is influenced standards and the literature refer to the weekly and by several indicators aimed at the signal transmission. yearly availability of the alarm transmission system, as Those indicators could include, for example, message shown in Table 1. The calculation and/or expression of transmission time, connection/connection monitoring, availability can be done using formulas to calculate the monitoring of the individual parts of the alarm transmission weekly, monthly and yearly availability values of the alarm system functioning [4]. transmission system level of table 1 for each class. The alarm transfer time is determined by the arithmetic Regular tests of connection of the protected object mean of the message transfer time and 95 percens of the and the alarm system monitoring center should be carried message transfer time measurement. By this combination, out to obtain the data necessary for expressing availability. one gets the maximum allowed time of the message Tests of this type are carried out by operators of the alarm transmission from the protected object to the alarm system monitoring center every week and are particularly transmission system center. The maximum allowable useful for their personal needs. For scientific research, it message transfer time values are shown in Table 1. The is necessary to carry out the comprehensive tests covering message transfer time is the period measured from the several parts of the object protection system such as the moment when the information appears on the interface electrical security system, the camera system and the like of the transmitting device located in the protected area [5-6]. to the time it is recorded at the interface of the receiving device located in the alarm transmission center system. Any message transfer time exceeding the maximum allowed 2 Methodology message transfer time, specified in Table 1, shall be defined as a transmission system failure. Within the scope and focus of the paper, the focus was set only on experimental testing of the alarm transmission

COMMUNICATIONS 2/2020 VOLUME 22 POSSIBILITIES FOR EXPERIMENTAL TESTING OF ALARM TRANSMISSION SYSTEMS 125

Figure 2 Block diagram of transmission path connection during experimental testing

Figure 3 Wiring diagram of the test device together with an example of real connection of the test equipment systems. The experiment, or the method of experiment, has and constructed to be compatible with transmission a key position among the methods of quantitative research equipment operating in different transmission networks. As because it is the only one able to prove or confirm causal can be seen in Figure 2, the GSM transmission network was consequences, respectively. Its implementation is mostly used during the experimental tests [7]. used to verify experience and to confirm or refute scientific The Arduino platform was chosen because of its questions or hypotheses. In the experiment, one encounters extensibility and the considerable amount of free independent variables that can be modified and changed information and knowledge from various online forums. depending on the requirements to make their impact on The test device is based on a programmable Arduino NANO results clear. motherboard, which is adapted to be used in a plug-in, Through the experimental testing of the alarm connection box without soldering. The testing device also transmission system, the scientific activities of the Faculty includes a real-time module and an SD card slot on which of Security Engineering focus on creating a database of data each change of the system status is recorded, an LED for designed to assess the reliability of the alarm transmission optical status change indication. The block diagram of the system. The focus was placed on the transmission test device is shown in Figure 3. information time, from the protected object to the alarm Communicator GC-20A is designed for installation transmission system center. In conducting the experimental in smaller buildings such as houses, cottages where tests, the unified connection of the alarm transmission path transmission of many messages is not expected. The was used, block diagram of which is shown in Figure 2. communicator uses the Contact ID (CID) encryption to The experimental testing was preceded by the creation of transmit information. The CID is a digital communication a test facility that can be considered as an innovative way of format that contains fifteen characters that together obtaining the data needed to evaluate the object protection form a description of the event with the identification system. of the location of the alarm in the protected object. The The test facility is built using components on the communicator can create a backup transmission path in open-source Arduino platform. Testing equipment during the form of SMS messages, as GPRS transmission is used experimental testing by connecting components and they as the primary one. The GSM Wireless Airlink Fastrack aimed to determine the real functionality and possibility of Xtend EDGE FXT009 GSM modem is a fully programmable use in practice. Therefore, at this stage of testing, the tester GSM modem that can be extended with GPS modules. did not have any type of box. The test equipment is designed The modem can be connected via traditional RS232 serial

VOLUME 22 COMMUNICATIONS 2/2020 126 BOROS et al.

Table 2 Measured values of experimental testing Property - 1 Property - 2 ATS Availability of the alarm transmission path [%] 90.90 88.10 - Availability of the alarm transmission system [%] - - 76.57

port and USB port. The Jablonet pro is a specialized 3 Results software enabling the management and development of alarm transmission systems. Thanks to its equipment it can Using the experimental tests, more than 1200 messages cooperate with many different types of monitoring devices. were transmitted from the protected object to the alarm Many experimental measurements were necessary to transmission system center to determine the average value of obtain the reliable data on reliability of the transmission the message time and availability of the alarm transmission systems. As a part of the research into the reliability of system. As mentioned above, to express the availability alarm transmission systems, in the first phase of testing, of the alarm transmission system, the calculation for the measurements from the two different protected objects, monthly availability of the alarm transmission system was in which no electrical security system was installed, used, which is expressed according to the relation: were carried out. The measurements were carried out SF simultaneously for 10 days from both objects to simulate MA =-1 ) 100 , (1) bl43800 the real use of the system. The basis of the measurements was a testing device from which information was sent to where: the alarm transmission system center at regular 20-minute MA - monthly expression of the availability of the alarm intervals about the intrusion of the protected object. In transmission system expressed as a percentage, all the cases of this phase of experimental testing, it was SF - the sum of breakdown times over 30 days, expressed the same type of alarm relay message in terms of size. in minutes, The reason was regular switching of the alarm input of 43 800 - the average number of minutes in one month the communicator, to which the unified type of event that except for a leap year. occurred during programming was set. Since this was Since in the pilot experimental tests measurements the first stage of experimental testing, focused on the lasted only 10 days, the relation for calculation was correct functionality of the test equipment, no restrictive adjusted. The adjustments consisted of changing the conditions for the time of transmission of the message were constant value of the average number of minutes per defined. The first phase was aimed to find out whether the month from 43 800 to 14 600, which is one-third, as the testing device can fully replace the control panel of the basis was on the SF parameter, which refers to a period of electric security system and thus simulate the protected 30 days. Using the adjusted formula, it was found that the object [8-9]. real availability of the alarm transmission system during The ten-day interval was chosen because of the ability the practical measurements was 76.57%. The availability of to modify the monthly availability formula. It was chosen the alarm transmission system is decisive for all the alarm to base the calculation on monthly availability because transmission routes through which the protected objects weekly availability is a verification calculation based are connected to the centralized security desk. on knowing annual availability. Since those were initial In evaluating the results of experimental testing, it tests, information on the annual availability of the alarm was also decided to determine the availability of the alarm transmission system could not have been provided. transmission path, i.e. one connected protected object. Given the possibilities and focus of experimental No attention was paid to this issue in terms of technical testing, the two different protected objects were chosen, literature and technical legislation and standards. Therefore, with different distances from the alarm transmission system a relationship was established allowing calculation of center. The object - 1 was approximately 10 kilometers from availability of the alarm transmission path, which is based the center of the alarm transmission system on the road and on the issue of system reliability and looks as follows: the air distance was approximately 6 kilometers. The object NM - 2 was approximately 16 kilometers from the center of the AATP =-1 ) 100 , (2) blM alarm transmission system on the road and the air distance was approximately 9 kilometers. In addition to different where: distances of the protected objects, there is also a difference APPC - Availability of alarm transmission path, in their location as the object - 1 is oriented to the southeast NM - Number of non-transmitted messages, and object - 2 to the south of the alarm transmission system M - Total number of messages. center. The center of the alarm transmission system is in Using Equation (2), availability of the alarm the premises of the Science Park of the University of Zilina. transmission path for each protected object was calculated

COMMUNICATIONS 2/2020 VOLUME 22 POSSIBILITIES FOR EXPERIMENTAL TESTING OF ALARM TRANSMISSION SYSTEMS 127

separately. In the case of the protected object 1, the one can talk about the quality of the transmission service availability of the alarm transmission path was 90.9% and in through a non-post channel. From the aforementioned the case of the protected object 2 88.1%. The experimental definition of the availability of an alarm transmission results obtained are also summarized in Table 2. system, it can be stated that it is precisely it from alarm As can be seen from results in Table 2, the availability transmission systems. In this statement one assumes that of the alarm transmission paths, as well as the alarm the whole testing methodology was designed and based on transmission system, is relatively low due to the short the issue of alarm transmission systems. testing time. It was found as very difficult to compare them The ten-day cycle that this paper is dedicated to with the requirements of the technical standards of Table was the initial phase of testing that was needed to create 1. The reason for the rather large distortion of the resulting the necessary image of the functionality of the whole, data was the malfunction of the receiving device from day presented alarm transmission system and testing equipment. 8 until the end of experimental testing on day 10. One could Currently, the second phase of testing is under way, which say that this is about 20% of all the data that affected the monitors the data transmission in the long term, while result. The reason for the malfunction was probably the removing the identified shortcomings from the first phase. absence of software used by the operator. The goal was to The data obtained so far are of a higher value in terms of automate the whole process, so there was no interference value and the expected completion of the second phase is with any part of the test. The equipment manufacturer in February 2020. was also informed of these findings, but no comment on Intruder alarm systems, video surveillance systems and the situation was possible to make, as this was a specific alarm transmission systems have been tested at the Faculty situation that could not occur under real conditions. The of Security Engineering. In terms of the effectiveness of center of the alarm transmission system is a permanently obtaining the data needed to evaluate the object protection operated workplace. system, it would be necessary to focus on experimental If, for example, on the second day of testing, the testing of the input control system and reliability of human receiving device malfunctions, the availability value will be factors to automate the object protection rating system at a diametrically different level. In this case, it would not even more accurately. be possible to evaluate the testing. From this point of view, one can state that the ten-day cycle was suitably chosen to point out the possible shortcomings of the set requirements 5 Conclusion following the next phase of experimental testing. The paper is aimed to point out a new possibility of obtaining the data necessary to re-evaluate the system of 4 Discussion object protection, which is formed by a combination of subsystems. The main topic was alarm systems, namely The need to test alarm systems has long been recognized alarm transmission systems, in which the first experimental in the circles of experts and scholars of literature and testing, focused on the availability of the transmission system should be given due consideration. The importance of using a testing device, was performed. When designing and experimental tests affects many people, such as security creating the test equipment, the simple programmable managers, private security operators, installers, academics circuits were used to simulate the connection of the and, finally, end-users [10]. protected object to the alarm transmission system center. In addition to users of individual systems, the results of Using the long-term testing, 10 days, it was possible experimental tests are also needed for simulation programs to determine the availability of the transmission system, as designed to evaluate the proposed systems. Those results well as the alarm transmission path, with minimal attention could influence design of the system at an early stage, to it. Experimental testing is based on the need to obtain the focusing on design efficiency and cost-effectiveness of the data needed for simulation programs to evaluate the object solution [11-12]. protection system. With selection of the alarm transmission Experimental testing was focused on an innovative way systems, the long-term scientific intention of the Faculty of testing alarm systems using the simple programmable of Security Engineering was followed upon, which, within circuits. Indeed, this type of testing can also be used for its scientific and publishing activities, devotes itself to the designing the critical infrastructure elements in different evaluation of object protection systems comprehensively. types of alarm systems. In addition to critical infrastructure, From individual parts of the object protection systems, the results of experimental tests can be used to ensure it is necessary to pay due attention to experimental testing the continuity of road traffic, particularly in the case of and to repeat it at regular intervals, focusing on analysis of video surveillance systems testing against extreme weather the reliability development of individual end parts. From conditions [13-14]. the comprehensive approach to evaluation of the object The results of the experimental tests can be understood protection system, the representatives of the Faculty of from several perspectives as it is an overlap of industry Security Engineering will deal with the systems of access interest. In terms of data transmission and capabilities, control.

VOLUME 22 COMMUNICATIONS 2/2020 128 BOROS et al.

References

[1] BOROS, M., SISER, A., KEKOVIC, Z., MAZAL, J. Mechanical characteristics of cylinder pin tumbler locks as they relate to resistance testing. Communications - Scientific Letters of the University of Zilina [online]. 2018, 20(2), p. 96-101. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/ view/94 [2] KITTEL, L., LOVECEK, T. Passive protection elements breach resistance modeling. Communications - Scientific Letters of the University of Zilina [online]. 2011, 13(2), p. 53-58. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/820 [3] BOROS, M., LENKO, F. Possibility of transmission system disruption by intruder. In: 13th International Scientific Conference on Sustainable, Modern and Safe Transport, TRANSCOM 2019 : proceedings [online]. 2019. ISSN 2352-1457. Available from: https://doi.org/10.1016/j.trpro.2019.07.176 [4] GUTTEN, M., JANURA, R., SEBOK, M., KORENCIAK, D., KUCERA, M. Measurement of short-circuit effects on transformer winding with SFRA method and impact test. Metrology and Measurement Systems [online]. 2016, 23(4), p. 521-529. ISSN 0860-8229. Available from: https://doi.org/10.1515/mms-2016-0044 [5] KUTAJ, M., JASENCAK, L., VELAS, A., BOROS, M. The design of a testing device for passive infrared motion detectors. WIT Transactions on The Built Environment, Safety and Security Engineering VII [online]. 2018, 174, p. 457-463. eISSN 1743-3509. Available from: https://doi.org/10.2495/SAFE170411 [6] VELAS, A., KUTAJ, M., DUROVEC, M. Influence of changing the parameters of the camera system on video-based motion detection. In: International Carnahan Conference on Security Technology ICCST 2017 : proceedings [online]. 2017, ISSN 1071-6572. Available from: https://doi.org/10.1109/CCST.2017.8167829 [7] MOZER, V., LOVECEK, T., VELAS, A., MAKOVICKA, L. Fire safety and security threats identification and elimination. Advanced Materials Research [online]. 2014, 1001, p. 306-311. ISSN 1022-6680. Available from: https://doi.org/10.4028/www.scientific.net/AMR.1001.306 [8] VELAS, A., LOVECEK, T., VALOUCH, J., DWORZECKI, J., VNENCAKOVA, E. Testing radio signal range of selected components. Communications - Scientific Letters of the University of Zilina [online]. 2018, 20(2), p.. 68-77. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/90 [9] URBANCOKOVA, H., KOVAR, S., HALASKA, O., VALOUCH, J., POSPISILIK, M. Conditions for testing effects of radio- frequency electromagnetic fields on electronic device. WSEAS Transactions on Environment and Development. 2018, 14, p. 145-152. ISSN 1790-5079, eISSN 2224-3496. [10] LOVECEK, T. Present and future ways of physical property protection. Communications - Scientific Letters of the University of Zilina [online]. 2008, 10(1), p. 35-39. ISSN 1335-4205, eISSN 2585-7878. Available from: http://komunikacie.uniza.sk/index.php/communications/article/view/1028 [11] RISTVEJ, J., ONDREJKA, R., SIMAK, L., LOVECEK, T., HOLLA, K., LACINAK, M., SURINOVA, L., JANOSIKOVA, M. Simulation technologies in risk prevention within crisis management. In: 30th European Simulation and Modelling Conference ESM 2016 : proceedings. 2016, ISBN 978-907738195-3. [12] KUBAS, J., SOLTES, V., MISIK, J., STOFKOVA, Z. Efficiency of using financial resources and their impact on security in a local context. In: 12th International Scientific Conference of Young Scientists on Sustainable, Modern and Safe Transport TRANSCOM 2017 : proceedings [online]. 2017, ISSN 1877-7058. Available from: https://doi.org/10.1016/j. proeng.2017.06.086 [13] SOLTES, V., KUBAS, J., STOFKOVA, Z. The safety of citizens in road transport as a factor of quality of life. In: 22nd International Scientific on Conference Transport Means : proceedings. 2018, ISSN: 1822-296X. [14] LOVECEK, T., SISER, A., MARIS, L. Use case of waterwork physical protection system robustness evaluation as a part of Slovak critical infrastructure. In: International Carnahan Conference on Security Technology ICCST 2017 : proceedings [online]. 2017, ISSN 10716572. Available from: https://doi.org/10.1109/CCST.2017.8167811

COMMUNICATIONS 2/2020 VOLUME 22 Editor-in-chief: Branislav HADZIMA - SK COMMUNICATIONS Associate editor: Jakub SOVIAR - SK VEDECKÉ LISTY ŽILINSKEJ UNIVERZITY SCIENTIFIC LETTERS OF THE UNIVERSITY OF ZILINA Executive editor: Sylvia DUNDEKOVA - SK VOLUME 22 Number 2 Honorary members: https://doi.org/10.26552/com.J.2020.2 Otakar BOKUVKA - SK Jan COREJ - SK (in memoriam) Milan DADO - SK Pavel POLEDNAK - CZ Author guidelines Editorial board: Greg BAKER - NZ • All papers have to deal with the topic of transport and be submitted Abdelhamid BOUCHAIR - FR Pavel BRANDSTETTER - CZ strictly within one of the listed subtopics. Please, refer to list of topics Mario CACCIATO - IT and subtopics here and indicate it clearly when submitting your paper. Jan CELKO - SK Andrew COLLINS - GB • Submitted papers must be unpublished and must not be currently under Samo DROBNE - SI review for any other publication. Erdogan H. EKIZ - MA Michal FRIVALDSKY - SK • Manuscripts written in good English must include abstract and keywords Juraj GERLICI - SK also written in English. The abstract should not exceed 10 lines. Vladimir N. GLAZKOV - RU Ivan GLESK - GB • Submitted manuscripts should not exceed 20 pages including figures Mario GUAGLIANO - IT and graphs Andrzej CHUDZIKIEWICZ - PL Jaroslav JANACEK - SK • Submission should be sent by e-mail – as an attachment – to the Zdenek KALA - CZ Antonin KAZDA - SK following address: [email protected]. Michal KOHANI - SK • The author‘s exact mailing address, full names, E-mail address, Jozef KOMACKA - SK Matyas KONIORCZYK - HU telephone or fax number, the name and address of the organization and Tomas LOVECEK - SK workplace (also written in English) must be enclosed. Frank MARKERT - DK Jaroslav MAZUREK - SK • For all manuscripts a double-blind peer review by at least two Marica MAZUREKOVA - SK independent reviewers and language correction is mandatory. Vladimir MOZER - CZ Jorge Carvalho PAIS - PT • After reviewing and incorporating the editor‘s comments, the final draft Peter POCTA - SK Maria Angeles Martin PRATS - ES (before printing) will be sent to authors for final review and minor Pavol RAFAJDUS - SK adjustments. Che-Jen SU - TW Giacomo SCELBA - IT Janka SESTAKOVA - SK The full author quidelines are available at: Eva SVENTEKOVA - SK Eva TILLOVA - SK http://komunikacie.uniza.sk/index.php/communications/guidelines Anna TOMOVA - SK Franco Bernelli ZAZZERA - IT

POKYNY PRE AUTOROV PRÍSPEVKOV DO ČASOPISU Address of the editorial office: University of Zilina KOMUNIKÁCIE – vedecké listy Žilinskej univerzity EDIS – Publishing House • Univerzitna 8215/1 • 1. Redakcia prijíma iba príspevky doteraz nepublikované alebo inde nezaslané 010 26 Zilina, Slovakia na uverejnen E-mail: [email protected] • • Individual issues of the journal can be found on: • http://komunikacie.uniza.sk • Each paper was reviewed by two reviewers. • • Journal is excerpted in SCOPUS and EBSCO. • • Published quarterly by University of Zilina in • EDIS – Publishing House of University of Zilina • Registered No: EV 3672/09 • 8. Príspevok posúdi redakčná rada na svojom najbližšom zasadnutí a v prípade jeho zaradenia do niektorého z budúcich čísiel podrobí rukopis recenziám a jazy- ISSN (print version) 1335-4205 ISSN (online version) 2585-7878 kovej korektúre. Pred tlačou bude poslaný autorovi na definitívnu kontrolu. • 9. Termíny na dodanie príspevkov do čí- ICO 00397 563 siel v roku sú: 30. september, 31. december, 31. marec a 30. jún. April 2020 • 10. Nosná téma pre č. 1/2008: Bezpečnostné inžinierstvo.